JPWO2009001696A1 - Information processing apparatus, program, and information processing method - Google Patents

Abstract

The information processing apparatus according to the present invention includes a storage unit 2 that stores document data, an input unit 3 that receives information on a document data group to be analyzed from a user, and a storage unit 2 based on the received information on the document data group. The document data group is obtained from the document data, morpheme analysis processing is performed on the document data group, each document data is decomposed into morpheme data, a feature word composed of each morpheme data is generated according to a predetermined rule, and the generated feature word is used. A control unit 10 for generating information indicating the tendency of each document data, and a display unit 4 for outputting information indicating the tendency of each document data. This provides an information processing apparatus that allows a user to guess the characteristics of a document more easily than before without decoding the document.

Description

  The present invention relates to a technique for analyzing document data, and more particularly to a technique for analyzing characteristics of document data and outputting an analysis result.

  Conventionally, Patent Document 1 and Patent Document 2 below are disclosed as techniques for analyzing the characteristics of document data.

  The technique disclosed in Patent Document 1 is intended to perform keyword extraction of document data at high speed, calculates the appearance frequency for all morphemes in the document data, and calculates the degree of coincidence with other morphemes This is a technique for extracting a keyword without performing a process such as.

  Specifically, a word corresponding to a noun that is led to a case particle or a particle is extracted as a keyword of the document data from among morphemes in the document data. This word is considered to be taken up as a topic in the document data, so that keywords are extracted from the document data at high speed.

  Further, the technique disclosed in Patent Document 2 is intended to extract and present a phrase so that the contents of the document can be sufficiently grasped, and extract an important phrase from document data. This is a technique for achieving the above-mentioned object by extracting a subject presentation word / phrase presented as the subject of the document data and presenting the subject presentation word / phrase and the important word / phrase in association with each other.

Specifically, the similarity between words is calculated for all the words in the document data, consecutive words attached to the auxiliary particles in the document data are extracted as the subject-presented phrases, and the words included in the subject-presented phrases A word having a high similarity is extracted as an important phrase, and the subject presentation phrase and the important phrase are linked and displayed. In this way, by displaying linked words and phrases that are closely related to the subject of the document, it is possible to make it easier for the user to understand the contents of the document than when the extracted phrases are simply displayed in the order of appearance.
JP 11-328206 A JP 2000-298673 A

  However, the above-mentioned patent document 2 merely links and outputs important words / phrases having high similarity with the main words / phrases of the document, and the user sequentially reads each word / phrase linked to grasp the relationship between the words / phrases. Then, the characteristics of the document must be estimated from the relationship between the phrases, which imposes a burden on the user.

  Furthermore, since each word output in Patent Document 2 is a morpheme after the morpheme analysis processing is performed, it is difficult to understand the meaning of the document by grasping the relationship between the words represented by the morpheme There is also a possibility. Moreover, even if the words and phrases are generated by applying the technique of the above-mentioned Patent Document 1, there is no change in the burden that the user sequentially reads each word and phrase linked to understand the meaning of the document.

  Therefore, the present invention has been made in view of the above situation, and an object of the present invention is to provide an information processing apparatus that allows a user to more easily infer a document feature than when a user understands the document feature. To do.

  In order to solve the above problems, an information processing apparatus according to the present invention performs morphological analysis processing on document data, detects morphemes in the document data, decomposes the document data into morpheme data, and converts the document data into An information processing apparatus for analyzing, a storage unit for storing the document data, and performing the morpheme analysis processing on the document data, and generating a first feature word composed of the morpheme data based on a predetermined first rule Feature word generating means, and output means for performing output processing of information indicating the tendency of the document data using the first feature word generated by the feature word generating means.

  With the above configuration, the information processing device according to the present invention generates the first feature word from the morpheme data of the decomposed document data according to the rule predetermined by the predetermined first rule. Since feature words can be routinely generated without performing processing such as document syntax analysis and semantic analysis, the processing load when generating feature words can be reduced. Further, since the first feature word generated using the morpheme data of the document data includes the contents of the document data, the information representing the tendency of the document data using the first feature word is included in the document. Characteristics and ideas are suggested. Therefore, the user can infer the contents and trends of the document data by referring to the information.

  Furthermore, the information processing apparatus according to the present invention further associates each decomposed morpheme data, a predetermined part of speech corresponding to each morpheme data, and detection order information indicating a detection order of each morpheme data. Part-of-speech information generating means for generating one part-of-speech information, wherein the first feature word includes the first part-of-speech for each predetermined case particle when the first part-of-speech information includes a predetermined case particle; Of the morpheme data of information, it may be composed of forward morpheme data which is morpheme data detected before the predetermined case particle.

  According to this configuration, the information processing apparatus can determine the presence / absence of the predetermined case particle and the detection position in the document data based on the first part of speech information generated by the part of speech information generation unit. Generally, words such as subject and purpose in a document are described before case particles such as “O”, “GA”, and “NO” in the document. Therefore, for each given case particle in the document, the first feature word composed of the forward morpheme data detected before that case particle contains morpheme data indicating the subject, purpose, etc. of the document The user can more easily grasp the contents of the document from the output first feature word.

  Further, the first rule is, based on the front morpheme data detected immediately before the predetermined case particle in the first part of speech information among the front morpheme data for each predetermined case particle in the first part of speech information. The forward morpheme data until the forward morpheme data belonging to parts of speech other than the first category are detected may be combined in the order of detection.

  According to this configuration, the first feature word can be generated simply by determining whether or not the part of speech corresponds to the predetermined part of speech using the part of speech given to the morpheme data by the conventional general morpheme analysis process. Therefore, the first feature word can be easily generated, and the processing load when generating the first feature word can be reduced.

  Further, the first classification may be a noun.

  The part-of-speech information generation means generates the part-of-speech information by using, as an unknown word, a part-of-speech of morpheme data that does not correspond to any of the predetermined part-of-speech among the decomposed morpheme data. The unknown word may be included.

  The first classification may further include symbols and adjectives.

  According to this configuration, since the noun described immediately before the predetermined case particle generally indicates the subject, purpose, object, etc. in the document, the user can detect the noun detected immediately before the predetermined case particle. From the first feature word generated by combining, the subject, idea, etc. in the document can be inferred. An unknown word is a coined word, technical term, or the like uniquely generated by a document creator, and an unknown word may be frequently used particularly when the document is not a general document. Therefore, when generating the first feature word, in addition to nouns, unknown words are to be combined, so even if the document is not a general content document, the user can more accurately identify the content and features of the document. I can grasp it.

  Symbols are used, for example, to represent numerical units and mathematical formulas used in documents, and often have important meanings in documents. Adjectives are the nature and state of the subject or target noun. It indicates the amount and degree. Therefore, by using these as the first feature word combination target, it is possible to make the user understand the contents of the document more clearly.

  Further, the document data may be patent document data including claim scope data described as claims, and the morpheme analysis process may process the claim scope data.

  Reading a patent document such as a patent application specification and claims and grasping the technical characteristics of each patent document requires a great deal of labor and burden even if the user is not familiar with the patent document. In particular, the content described in the claims is the part that represents the most characteristic technical idea of the patent document, and it is very difficult to read the description and immediately derive the characteristic part. It is. According to this configuration, among the morpheme data extracted from the claim data, the first feature word can be easily obtained by using the morpheme data indicating the subject or object of the technical idea claimed in the claim. The user can infer the subject technical feature of the patent document from the first feature word.

  The output means determines an output rank of the first feature word based on the appearance frequency of the morpheme data constituting the first feature word, and outputs the first feature word according to the output rank. It is good also as performing the output process of the information which shows the said tendency.

  According to this configuration, when the output unit outputs the first feature word, the output order is determined based on the appearance frequency of the morpheme data constituting the first feature word, and the output is performed according to the determined output order. The first feature word is output in a state where a certain order is maintained. Therefore, the user can easily estimate the position of the first feature word in the document and the relationship between the first feature words from the output state of the first feature word, and simply output the keywords in the document side by side. Compared to the case, the characteristic tendency of the document can be grasped more easily.

  Further, the information processing apparatus further determines whether or not the description format of the claim range data matches a predetermined format, and when determining that the description format matches the predetermined format, The data may include a specifying unit that specifies data in a range between predetermined data predetermined by the predetermined format as the morphological analysis processing target.

  According to this configuration, for example, there may be a case where not a characteristic wording in a patent document but a very standard wording is included in describing the scope of claims. It is possible to determine whether or not the claims are described in a predetermined format including the following words. For this reason, it is possible to exclude in advance the data of a fixed phrase that is not a feature part included in the claims from the analysis target, and it is possible to improve the accuracy in generating the first feature word.

  The patent document data is a plurality of patent document data, and each patent document data further includes problem information including a plurality of character string data as a problem of the document. And further comprising cluster classification means for classifying the plurality of patent document data into clusters based on the task information of each patent document data, wherein the feature word generation means further includes forward morpheme data of the predetermined case particles. Among them, the front morpheme data whose part of speech belongs to the first classification is extracted as solution word candidate target data in the claim data group of the plurality of document data, and the plurality of patent documents of the respective solution word candidate target data A solution word extraction means for calculating an appearance frequency in the data and extracting solution candidate object data having a predetermined rank or higher in descending order of the calculated appearance frequency as a solution word; A solution word appearance frequency calculating unit that calculates an appearance frequency of each solution word in each cluster, and the first feature word includes the solution word but has an appearance frequency in the plurality of patent document data. The appearance frequency of the morpheme data that satisfies a predetermined criterion and that constitutes the first feature word is the appearance frequency calculated by the solution word extraction unit of the solution word included in the first feature word The output unit may further output the solution word corresponding to the first feature word and the appearance frequency of the solution word calculated by the solution word appearance frequency calculation unit for each cluster.

  According to this configuration, the first feature word is morpheme data whose frequency of appearance in a plurality of patent document data groups is a certain level or more among morpheme data in which the part of speech detected before a predetermined case particle belongs to the first classification. Since a certain solution word is included, a general or useful technical word can be inferred from the first feature word in the entire patent document data group.

  Further, the patent document data group can be classified into a plurality of clusters by the cluster classification means, and the output means is provided for each cluster, each first feature word including each solution word and each solution word, and each solution word. Since the appearance frequency in the cluster is output in association with each other, the user can grasp the general technology and characteristic technology for each cluster by referring to the appearance frequency for each cluster corresponding to each solution word and the first feature word. can do. Furthermore, each solution word and each first feature word including each solution word are output according to the rank based on the appearance frequency in the patent document data group of each solution word by the output means, for example, in order of appearance frequency When output side by side, the user can easily confirm the technical tendency of the entire patent document data group from the output position of the solution word and the first feature word.

  The information processing apparatus further performs a morpheme analysis process on each piece of task information, and the part-of-speech information generation unit further includes each morpheme data decomposed by the morpheme analysis process on the task information, and each morpheme data. Second part-of-speech information that associates a predetermined part-of-speech corresponding to the morpheme data with detection order information indicating the detection order of each morpheme data, Calculates the appearance frequency in each patent document data for the morpheme data belonging to the first classification, generates a document vector of each document data using each calculated appearance frequency as an element, calculates the similarity between the document vectors, and A plurality of patent document data is classified into clusters, and the detection order in each patent document data of each morpheme data whose part of speech belongs to the first classification Specify based on the second part-of-speech information, generate combined morpheme data by combining the identified morpheme data having consecutive detection orders, calculate the appearance frequency in each cluster of the generated combined morpheme data, and A task word extracting means for extracting each combined morpheme data of a predetermined order or more in descending order as a task word indicating the cluster, and the output means further outputs the task words indicating the clusters. Also good.

  The problem information of each patent document data indicates a technical problem motivated when the technical idea claimed by each patent document is derived. According to this structure, the subject common to the patent document data group which belongs to each cluster can be easily extracted based on a fixed rule by a subject word extraction means. Further, the output means outputs each solution word and each first feature word, the appearance frequency of each solution word in each cluster, and the task word indicating each cluster, so that the user can select the task word. By referencing, differences in technical means and similarities for each problem can be compared.

  Further, the output means represents each task word and each solution word and the set of the first characteristic words corresponding to the solution word in each row and each column of the matrix, and each matrix As an element, it is good also as representing the value of the appearance frequency of the said solution word corresponding to the subject word and solution word which the said row and column which comprise the said element show.

  According to this configuration, the output unit outputs each task word, the first feature word including each solution word and each solution word, and the appearance frequency value in each cluster of each solution word on the matrix. The user can confirm the tendency of the technical means for each task word at a glance.

  The output means further generates a cluster document vector indicating the document vector of the cluster using the document vector of each document data generated by the cluster classification means, and uses the cluster document vector of each cluster. Then, by calculating the distance between the clusters, the intercluster similarity calculating means for obtaining the similarity between the clusters, and the value of the appearance frequency for each solution word calculated by the solution word frequency calculating means A solution word similarity calculating means for calculating a distance between the solution words by calculating a distance between the solution words using the vector of each solution word, For each similarity calculated by the similarity calculation means and the solution word similarity calculation means, a combination having the maximum similarity is extracted, and the extracted class And a cluster having each element included in each set of solution words as a reference element and each reference link set of solution words, respectively, and the one of the two connection directions starting from the reference element of the reference link The connection direction is set to one reference connection of each set of the standard connection of the cluster and the solution word, the other connection direction is set to the other reference connection of the set, and the similarity of each of the clusters and the solution words is determined. A cluster set and a solution word set are sequentially selected in descending order, and according to a predetermined condition, a connection process is performed to connect the elements of each selected combination to each of the reference connections. And an output order determining means for determining an output order of the solution words, wherein the predetermined condition includes a first condition and a second condition, and the first condition is a combination of the selected cluster and solution word. In One element included in the combination is connected to any reference connection, and the second condition is that any element included in the combination is unconnected, and the connection process includes: When any element of the selected combination satisfies the first condition, the unconnected elements of the combination are connected in the connection direction set in the standard connection including the connected elements satisfying the first condition. And a first connection process for generating a new reference connection set using each element of the combination as a new reference element when each element of the selected combination satisfies the second condition; A third connection process for connecting the reference connections of each set according to the number of elements connected to each of the reference connections, and connecting the connected reference connections of each set based on a predetermined rule, The output is the output order determining means. Thus, according to the determined output order of each cluster and each solution word, in each row and each column of the matrix, each task word and each solution word and a set of first characteristic words corresponding to the solution word are represented. It may be done.

  According to this configuration, the similarity between clusters and each solution word are calculated by the cluster similarity calculation means and the solution word similarity calculation means, and are selected in descending order of similarity by the output order determination means. The elements of the cluster and the solution word are connected in the same connection as the elements having high similarity to the elements. In addition, the cluster and the solution word are output for each connection having a certain relationship in each axis indicated by the matrix row example. In each connection, the cluster and the solution word are similar to the reference element having the highest similarity. The elements are output in order from the highest. Therefore, the user can confirm the distribution status of the appearance frequency of the solution word in the entire patent document data group in a set unit of clusters and solution words having a certain relationship on the matrix. For example, when a color is given to each cell according to the value of the appearance frequency output to each cell on the matrix, the user can check the distribution status of the appearance frequency at a glance, Intuitively understand the technical characteristics of the problem.

  In addition, the feature word generation means is further detected between each predetermined case particle for each predetermined case particle when the predetermined part particle is included in the first part of speech information. Generating, based on a second rule, a second feature word corresponding to the first feature word composed of backward morpheme data which is morpheme data detected after the predetermined case particle among the morpheme data; May output the second feature word corresponding to the first feature word together with the first feature word.

  According to this configuration, the second feature word corresponding to the first feature word is generated from the rear morpheme data after the predetermined case particle used for generating the first feature word by the feature word generation unit. Since the back morpheme data includes a morpheme representing the state or action of the subject or object described immediately before the predetermined case particle, the user can use the second feature word corresponding to the first feature word. The semantic content of the first feature word can be complemented, and the document content can be easily grasped.

  In addition, the second rule is that a back morpheme whose part of speech belongs to a part of speech other than the second classification from back morpheme data detected immediately after the predetermined case particle among the back morpheme data for each of the predetermined case particles. The backward morpheme data until the data is detected may be combined in the detection order.

  According to this configuration, the second feature word can be generated simply by determining whether or not the part of speech corresponds to the predetermined part of speech using the part of speech given to the morpheme data by the conventional general morpheme analysis process. Therefore, the second feature word can be easily generated, and the processing load when generating the second feature word can be reduced.

  The second classification may be a noun and a verb.

  According to this configuration, the feature word generation unit combines the back morpheme data of the noun or verb that continues from immediately after the predetermined case particle as the second feature word. In general, immediately after a given case particle, a word having a deep relationship with the word described immediately before the given case particle is described, and nouns and verbs are present in the word immediately before the given case particle. Therefore, it is easier for the user to understand the meaning and document content of the first feature word from the first feature word and the second feature word corresponding to the first feature word.

  The document data is a plurality of document data, and the output means further includes an appearance frequency calculating means for calculating an appearance frequency in the plurality of document data for each front morpheme data of the plurality of document data. And the appearance frequency of the morpheme data constituting the first feature word is an appearance frequency calculated by the appearance frequency calculation means of each forward morpheme data constituting the first feature word, and each document data The output of the first feature word and the second feature word corresponding to the first feature word may be performed in the order of the appearance frequency of the front morpheme data of each first feature word.

  According to this configuration, the first feature word is generated by combining the front morpheme data for each predetermined case particle in the document data by the feature word generation unit, and is a general term in the document data group. Since it cannot be said that it has been described, a difference is unlikely to occur in the appearance frequency of each first feature word in the document data group. On the other hand, since the forward morpheme data constituting the first feature word is easily used as a general term in the document data group, a difference in appearance frequency in the document data group of each forward morpheme data is likely to occur. Therefore, by outputting each first feature word using the appearance frequency of the forward morpheme data constituting the first feature word, each first feature word can be ordered and output for each document data. Therefore, the user can confirm the position of each feature word in each document data at a glance from each first feature word and second feature word output in order, and the characteristic tendency of each document data Can be grasped.

  Further, the information processing apparatus further calculates an appearance frequency in each document data for the front morpheme data constituting the first feature word among the respective front morpheme data of the plurality of document data, and the calculated appearance A document vector having frequency values as elements is generated, and using each document vector, a similarity calculation unit that calculates a similarity between each of the plurality of document data, and each calculated by the similarity calculation unit As for the similarity, each combination having the maximum similarity is extracted, each reference data set included in each extracted set is generated as each reference document data, and the reference connection reference document data is used as the starting point. One of the two connection directions is set as one reference connection of a set of reference connections, the other connection direction is set as the other reference connection of the set, and the documents are arranged in descending order of the similarity. Decision to select a set of data sequentially, perform a concatenation process to concatenate each selected set of document data to each of the reference concatenations according to a predetermined condition, and determine the output order of all the document data according to the result of each concatenation process And the predetermined condition includes a first condition and a second condition, and the first condition is that each document data of each combination is linked to any one of the standard connections. The second condition is that any document data is unconnected, and the connection process is performed when any one of the selected combinations of document data satisfies the first condition. , The first connection process for connecting the unconnected document data of the combination in the connection direction set in the reference connection including the connected document data satisfying the first condition, and the document data of both of the selected combinations When the second condition is satisfied, the second connection processing for generating a new reference connection set using both the document data as new reference document data, and the number of document data connected to each reference connection And a third connection process for connecting the reference links of each set according to each other and connecting the connected reference links of each set based on a predetermined rule, wherein the output means is each document determined by the determining means. The first feature word and the second feature word of each document data may be output for each document data according to the order of data.

  According to this configuration, the similarity between the document data is calculated by the similarity calculation unit, and each document data selected by the determination unit in descending order of the similarity is linked to the same connection as the document data having high similarity. In each connection, the document data are connected in the descending order of the similarity with respect to the reference document data having the highest similarity. Therefore, the first feature word and the second feature word of each document data are output so that a document data group having a certain relationship with the document data can be understood. The burden of grasping can be reduced.

An information processing method according to the present invention is a document analysis method that performs morpheme analysis processing on document data, detects morphemes in the document data, decomposes the document data into morpheme data, and analyzes the document data. , A feature word generation step of generating a first feature word composed of the morpheme data based on a predetermined first rule, and the first feature word based on the appearance frequency of the morpheme data constituting the first feature word And an output step of outputting the first feature word according to the output order.
The program according to the present invention is a program that is executed by an information processing apparatus including a CPU, performs morpheme analysis processing on document data, detects morpheme in the document data, and decomposes the document data into morpheme data. And a feature word generating step for generating a first feature word composed of the morpheme data based on a predetermined first rule, and a first feature word based on an appearance frequency of the morpheme data constituting the first feature word. An output step of determining the output order of one feature word and outputting the first feature word according to the output order.

  With the above configuration, the information processing method and the program according to the present invention generate the first feature word from the morpheme data of the decomposed document data according to the rule predetermined by the predetermined first rule in the feature word generation step. As a result, feature words can be generated in a fixed manner without performing processing such as document syntax analysis and semantic analysis, thereby reducing the processing load when generating feature words. Furthermore, when outputting the first feature word in the output step, the output order is determined based on the appearance frequency of the morpheme data constituting the first feature word, and output according to the determined output order is performed. Words are output in a certain order. Therefore, the user can easily estimate the position of the first feature word in the document and the relationship between the first feature words from the output state of the first feature word, and simply output the keywords in the document side by side. Compared to the case, the characteristic tendency of the document can be grasped more easily.

  Further, in the information processing apparatus, the storage unit stores a plurality of the patent document data, and the feature word generation unit includes the patent document data in the claim data of each patent document data. The first feature word is generated using the morpheme data of the first predetermined part including the character string indicating the technical feature constituting the invention, and the patent document in the claim data of each patent document data A third feature word is generated using the morpheme data of a second predetermined portion including a character string indicating the subject of data invention, and the information processing apparatus further includes the morpheme data included in each third feature word The plurality of patent document data is clustered using the first appearance frequency in the plurality of patent document data, and each of the patent document data corresponding to each of the third feature words belongs. Generating a document vector of each patent document data based on a second appearance frequency in the plurality of patent document data of each of the first feature words, and using each document vector Based on the factor loading, factor analysis means for performing factor analysis using the first feature words as observation variables, calculating factor loadings of the first feature words and factor scores of the patent document data, and Factor specifying means for specifying the factor of each first characteristic word, specifying the factor of each patent document data based on the factor score, and the first corresponding to each factor specified by the factor specifying means A technical element keyword indicating the factor is generated using the feature word, and the third feature word of the patent document data belonging to each cluster specified by the cluster specifying means is used to generate the keyword. Keyword generating means for generating a product group keyword indicating a star, and the output means is based on the factor of each patent document data specified by the factor specifying means as information indicating the tendency of the plurality of patent document data The relation information indicating the relation between each technical element keyword and each product group keyword may be output.

According to this configuration, the information processing apparatus according to the present invention performs the factor analysis of the patent document data group using the appearance frequency of the first feature word by the factor analysis unit, without requiring analogy by the user. It is possible to clarify the elements that are latent in the patent document data group, and to express each factor by a technical element keyword using the first feature word. Further, clustering of the patent document data group using the third feature word corresponding to each patent document data is performed by the cluster specifying means without preparing teacher data as a classification condition for classifying the patent document data group in advance. It can be performed with high accuracy and can be represented by a product group keyword using a third feature word for each cluster. Both the first feature word and the third feature word are generated for the claim data in which the technical scope of the invention of the patent document data is described. The first feature word is included in the patent document data group. Each of the third characteristic words corresponding to each patent document data represents an object of the invention of each patent document data.
Therefore, the user is latent in the patent document data group by the technical element keyword generated using the first characteristic word representing the technical element and the product group keyword generated using the third characteristic word representing the subject of the invention. Since it is possible to confirm the products and the like for which the invention of the technology and patent document data group is used, it is possible to grasp the tendency of the technology or product targeted by the patent document data group.
Further, the information processing apparatus according to the present invention can output relationship information indicating the relationship between each technical element keyword and each product group keyword based on factors of each patent document data. Each technical element keyword composed of the first feature word indicates a factor, and each product group keyword composed of the third feature word corresponds to each cluster. Therefore, the user can confirm the relationship between the technology latent in the patent document data group and the product in which each technology is used by the relationship information.

  In the information processing apparatus, the claim range data includes claim data for each claim, and the feature word generation unit generates the first feature word when the first feature word is generated. When the morpheme data of the first predetermined portion of each claim data in the claim data is used to generate the third feature word, the predetermined in the claim data of the patent document data The morpheme data of the second predetermined portion of the claim data may be used.

  According to this configuration, since the first feature word targets the first predetermined portion of all the claim data in the claim data of each patent document data, all the inventions included in the patent document data group The configured technical elements can be extracted. The third feature word indicates the subject of the invention of each patent document data, and in the description of each claim data, the word indicating the subject of the invention is often included in the same description location. Therefore, the processing load for generating the third feature word can be reduced by generating the third feature word using only the morpheme data of the second predetermined portion in the specific claim data of each patent document data. The object of the invention relating to each patent document data can be easily extracted.

  Further, in the information processing apparatus, the factor specifying means uses, as a factor of the first feature word, a factor in which the factor load amount of each first feature word calculated by the factor analyzing means is equal to or greater than a first threshold value. It is also possible to identify and identify as a factor of the patent document data a factor whose factor score of each patent document data calculated by the factor analysis means is a second threshold value or more.

  According to this configuration, since a factor that has a certain influence on each first feature word is specified as a factor of the first feature word, a technology closely related to the technical elements included in the patent document data group is specified. be able to. Further, since a factor having a certain contribution level for each patent document data is specified as a factor of the patent document data, a technique highly relevant to the invention of each patent document data can be specified.

  Further, in the information processing apparatus, the clustering by the cluster specifying means is performed based on a third appearance frequency in each third feature word of each morpheme data of the second predetermined portion, and the document vector of each third feature word And calculating a similarity between the document vectors of the third feature word having a fourth appearance frequency in the plurality of patent document data of each third feature word equal to or higher than a predetermined value, and according to the similarity A process of extracting a cluster, a similarity between the third feature word and the cluster having the fourth appearance frequency smaller than the predetermined value is calculated, and a patent document of the third feature word according to the similarity A process of including data in the cluster.

  According to this configuration, clusters are extracted by excluding third feature words in which the third appearance frequency of the third feature words in the patent document data group is smaller than a predetermined value, and a cluster having high similarity with the third feature words is extracted. Since the third feature word is included, a large number of small clusters can be prevented from being extracted, and useful clusters can be extracted from the patent document data group.

  Moreover, in the information processing apparatus, the keyword generation unit includes a factor loading amount of the factor that is greater than or equal to a third threshold among the first feature words corresponding to the factors identified by the factor identification unit. The technical element keyword is generated by combining the first feature words, and for each cluster extracted by the cluster specifying means, the centroid vector of the cluster and the third feature word of the patent document data belonging to the cluster The product group keyword may be generated by calculating a similarity with the document vector and combining the third feature words of the patent document data belonging to the cluster according to the similarity.

  According to this configuration, by combining only the first feature words having a factor load equal to or greater than a certain value among the first feature words corresponding to the factor, and generating the technical element keyword indicating the factor, Since only the first feature words whose descriptive power is above a certain level can be combined, a more appropriate technical element keyword can be generated as an expression indicating the factor. Further, by combining the third feature word of the patent document data according to the similarity between the cluster centroid vector and the patent document data of the cluster, and generating a product group keyword indicating the cluster, Thus, only the third feature word of more general patent document data can be combined. That is, a more appropriate product group keyword can be generated as an expression indicating the cluster.

  Further, in the information processing apparatus, the output unit counts the number of cases for each factor of the patent document data belonging to the cluster corresponding to the product group keyword for each product group keyword, and as the relation information, It is good also as outputting the information which matched the number of cases for each said factor of each said product group keyword, and the technical element keyword which shows the said factor.

  According to this configuration, the output means can output the number of patent document data belonging to the product group keyword using the technical element keyword as the relation information between the technical element keyword and the product group keyword in the patent document data group. it can. Therefore, for example, by referring to the related information, the user can confirm how much the technology that is latent in the patent document data group of a certain company is used for which product. It is possible to know whether or not duplicate research and development is being conducted.

  In the information processing apparatus, the storage unit further stores evaluation values corresponding to the plurality of patent document data, and the output unit stores the product group keyword for each product group keyword. The evaluation value of each patent document data belonging to the corresponding cluster is totaled for each factor, and as the relation information, the evaluation result for each factor of each product group keyword and a technical element indicating the factor It is good also as outputting the information which matched the keyword.

  According to this configuration, the output means can output the evaluation value aggregation of the invention belonging to the product group keyword related to the technical element keyword as the relation information between the technical element keyword and the product group keyword in the patent document data group. . Therefore, for example, when the evaluation value for each patent document data represents the importance of the invention related to the patent document data, for each technology included in the patent document data group, the technology is important in which product. It is possible to confirm which of the technologies used for each product is important.

The functional block diagram of the information processing apparatus which concerns on embodiment is shown. (a) is a figure which shows the structure and example of data of the part-of-speech information according to Embodiment 1 (1) table. (b) is a figure which shows the structure and data example of a document-specific part-of-speech information (2) table which concern on Embodiment 1. FIG. (c) is a figure which shows the structure and data example of a part of speech information (1) table which concern on Embodiment 1. FIG. (d) is a figure which shows the structure and example of data of the subject document vector information which concern on Embodiment 1. FIG. (a) is a figure which shows the structure and data example of cluster information which concern on Embodiment 1. FIG. (b) is a figure which shows the structure and data example of combined morpheme (2) information which concern on Embodiment 1. FIG. (c) is a figure which shows the structure and example of data of a subject word extraction list which concern on Embodiment 1. FIG. (a) is a figure which shows the structure and data example of solution word candidate morpheme information which concern on Embodiment 1. FIG. (b) is a figure which shows the structure and data example of a solution word list which concern on Embodiment 1. FIG. (c) is a figure which shows the structure and data example of combined morpheme (1) information which concern on Embodiment 1. FIG. (a) is a figure which shows the structure and example of data of the combination morpheme classified by solution word (1) information which concerns on Embodiment 1. FIG. (b) is a figure which shows the structure of the feature word extraction information according to solution word which concerns on Embodiment 1, and a data example. (a) is a figure which shows the structure and data example of a matrix element table which concern on Embodiment 1. FIG. (b) has shown the structure and data example of the solution word similarity information which concern on Embodiment 1. FIG. (a) is a figure which shows the example of the similarity between clusters for demonstrating the output order determination process which concerns on Embodiment 1. FIG. (b) is a figure which shows the result of having extracted the combination of the similarity degree shown to (a). It is a figure which shows the process of producing | generating the connection list | wrist of the output order determination process which concerns on Embodiment 1. FIG. 5 is a diagram illustrating an output example of a problem / solution matrix according to Embodiment 1. FIG. 3 is a diagram showing an operation flow of the information processing apparatus 1 according to Embodiment 1. FIG. 3 is a diagram showing an operation flow of the information processing apparatus 1 according to Embodiment 1. FIG. (a) is the figure which illustrated the solution information which concerns on Embodiment 1. FIG. (B) It is a figure which shows the analysis object extraction process flow which concerns on Embodiment 1. FIG. 6 is a diagram illustrating a clustering process flow according to Embodiment 1. FIG. It is a figure which shows the subject word extraction processing flow which concerns on Embodiment 1. FIG. FIG. 5 is a diagram showing a morpheme combining process flow according to the first embodiment. It is a figure which shows the extraction processing flow of the solution word and the feature word which concern on Embodiment 1. FIG. FIG. 6 is a diagram illustrating a feature word determination processing flow according to the first embodiment. FIG. 9 is a diagram showing a cluster output order determination processing flow according to the first embodiment; It is a figure which shows the output order determination processing flow concerning Embodiment 1 and Embodiment 2. FIG. It is a figure which shows the linked list order | rank determination processing flow which concerns on Embodiment 1 and Embodiment 2. FIG. FIG. 10 is a diagram showing a processing word output order determination processing flow according to Embodiment 1; FIG. 6 is a diagram illustrating a problem / solution matrix output process flow according to the first embodiment. It is a figure which shows the structure and data example of a part-by-document part of speech information table which concern on Embodiment 2. FIG. It is a figure which shows the structure of the feature word extraction list according to document which concerns on Embodiment 2, and a data example. (a) is a figure which shows the structure of the TF value table classified by document based on Embodiment 2, and a data example. (b) is a figure which shows the structure and example of a data of the GF value table classified by document based on Embodiment 2. FIG. (c) is a figure which shows the structure and data example of the output order information classified by document based on Embodiment 2. FIG. It is a figure which shows the example of an output of the feature word map which concerns on Embodiment 2. FIG. FIG. 10 is a diagram illustrating an operation flow of the information processing apparatus according to the second embodiment. FIG. 10 is a diagram showing a feature word extraction processing flow according to the second embodiment. FIG. 10 is a diagram showing a process flow for determining the output order of patent document data according to the second embodiment. It is a figure which shows the output order determination processing flow of the feature word information which concerns on Embodiment 2. FIG. It is a figure which shows the feature word map output processing flow which concerns on Embodiment 2. FIG. 6 is a diagram illustrating a functional configuration of an information processing device according to Embodiment 3. FIG. (a) shows the configuration and data example of the patent document data table in the third embodiment, and (b) shows the configuration and data example of the part-of-speech information table by application number in the third embodiment. (a) shows the configuration and data example of document vector information by technical element subject word in Embodiment 3, and (b) shows the configuration and data example of document vector information by application number in Embodiment 3. Show. (a) shows an example of claim data in the third embodiment, (b) shows a configuration and data example of factor load amount calculation result information in the third embodiment, and (c) shows an implementation. The structure of factor score calculation result information in the form 3 and the example of data are shown. (a) shows the configuration and data example of application number-specific attribution information in Embodiment 3, (b) shows the configuration and data example of technical element keyword information in Embodiment 3, (c) The structure of the product group keyword information in Embodiment 3, and the example of data are shown. (a) shows the configuration and data example of the cluster-specific factor number information in the third embodiment, and (b) shows the configuration and data example of the cluster-specific factor evaluation value information in the third embodiment. . 10 shows an operation flow showing the overall operation of the information processing apparatus 100 according to Embodiment 3. 10 shows a morphological analysis process flow according to the third embodiment. The product group object word production | generation processing flow which concerns on Embodiment 3 is shown. 10 illustrates a clustering process flow according to Embodiment 3. The factor analysis processing flow which concerns on Embodiment 3 is shown. The factor specific processing flow which concerns on Embodiment 3 is shown. 10 shows a keyword generation processing flow according to the third embodiment. 10 shows a related information output processing flow according to the third embodiment. (a) shows an output example of the first relation information according to Embodiment 3, and (b) shows an output example of the second relation information. 14 is a flowchart illustrating a procedure of cluster score calculation processing according to the third embodiment. The figure which simulated an example of the data structure of the progress information utilized by the calculation process of the patent score in Embodiment 3. FIG. The figure which simulated an example of the data structure of the content information utilized by the calculation process of the patent score in Embodiment 3. FIG. 10 is a flowchart illustrating a procedure of a patent score calculation process in the third embodiment. 10 is a flowchart showing details of processing for calculating an evaluation value of each patent data in the third embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1,100 Information processing apparatus 2 Memory | storage part 3 Input part 4 Display part 10,110 Control part 101 Input reception part 102 Data acquisition part 103,111 Morphological analysis part 104 Cluster analysis part 105,112 Feature word extraction part 106 Solution word extraction part 107 task word extraction unit 108 map generation unit 109, 117 output control unit 113 factor analysis unit 114 factor identification unit 115 cluster identification unit 116 keyword generation unit

[Embodiment 1]

<Overview>
The information processing apparatus according to the present embodiment analyzes the characteristics of a plurality of document data groups and outputs the analysis results.

  Specifically, the information processing apparatus according to the present embodiment generates a feature word indicating the characteristics of a plurality of document data groups using words in the plurality of document data according to a predetermined rule. In addition, the information processing apparatus classifies a plurality of document data groups into a plurality of clusters, ranks information indicating each cluster and each feature word, and displays them on each axis of the matrix rows and columns, respectively. In each cell indicated by the matrix, the value of the appearance frequency in each cluster of words that meet the predetermined condition constituting each feature word is displayed.

  In this embodiment, it is assumed that the plurality of document data is patent application data filed with the Japan Patent Office, and the user designates a technical field such as International Patent Classification (IPC). It is assumed that it is patent document data of a specific technical field extracted from patent application data by acceptance.

Each patent document data includes claims and summary data, and bibliographic data such as application date and applicant name.
Hereinafter, details of the information processing apparatus according to the above-described embodiment will be described.

<Configuration>
A configuration of the information processing apparatus according to the present embodiment will be described.

FIG. 1 is a diagram illustrating a functional configuration of the information processing apparatus according to the present embodiment.
In FIG. 1, the information processing apparatus 1 includes a storage unit 2, an input unit 3, a display unit 4, and a control unit 10.

  Hereinafter, the function of each part will be described in detail.

  The storage unit 2 is a recording medium such as a hard disk or a CD-ROM (Compact Disc Read Only Memory), and has a function of storing patent application data, data generated by each process by the information processing apparatus 1, and the like.

  The input unit 3 is realized by a keyboard, a mouse, or the like, and has a function of receiving an instruction to the information processing apparatus 1 such as designation of a technical field by a user.

  The display unit 4 is a display device such as a CRT (Cathode Ray Tube) display or a liquid crystal display, and has a function of displaying an image for accepting designation of a technical field from a user, an image of the matrix, and the like.

  The control unit 10 is realized by a CPU (Central Processing Unit) and a memory (ROM (Read Only Memory) and RAM (Random Access Memory)), and the CPU executes a program stored in the ROM to process information. It has a function of controlling each part of the device 1.

  The control unit 10 includes an input reception unit 101, a data acquisition unit 102, a morpheme analysis unit 103, a cluster analysis unit 104, a feature word extraction unit 105, a solution word extraction unit 106, a task word extraction unit 107, a map generation unit 108, and an output. A control unit 109 is included.

  Hereinafter, each part of the control part 10 is demonstrated.

  The input receiving unit 101 has a function of receiving an instruction from the user via the input unit 3 and, when the received instruction is instruction information indicating a technical field of document data, sending the instruction information to the data acquisition unit 102. Have.

  The data acquisition unit 102 extracts patent application data (hereinafter referred to as “designated patent document data group”) indicated by the instruction information received from the input receiving unit 101 from the storage unit 2 and is included in the designated patent document data group. Of the summary data, the data of the part described as “issue” (hereinafter referred to as “issue information”) and the data of claims (hereinafter referred to as “claim data”). Is sent to the morphological analysis unit 103.

  The morpheme analysis unit 103 performs morpheme analysis on each of the problem information and each claim data of each patent document data received from the data acquisition unit 102, and as a morpheme analysis processing result of the claim data, a patent document A document-specific part-of-speech information (1) table in which an identification number indicating the detection order of morpheme data, morpheme data, and the part of speech to which the morpheme data belongs is generated for each data. In addition, as a result of morphological analysis of the task information, similarly to the document-specific part-of-speech information (1) table, a document-specific part-of-speech information (2) table is generated for each patent document data, and each document-specific part-of-speech information (2) table It has a function of integrating the data and generating a part-of-speech information (2) table of the designated patent data group.

  It is assumed that grammar information used when performing morphological analysis and word list information associated with parts of speech are stored in advance in the information processing apparatus 1. In the present embodiment, when morphological analysis processing is performed on the claim range data, the morpheme analysis unit 103 performs analysis target extraction processing on the analysis target portion that performs the morphological analysis processing on the claim range data.

  Here, the analysis object extraction process extracts data described as “Claim 1” (hereinafter referred to as “solution information”) from the claims data of each patent document data, and provides the solution information. This is a process of determining whether or not the description format matches a predetermined format and extracting partial data to be subjected to morphological analysis in the solution information according to the determination result.

  The cluster analysis unit 104 has a function of performing cluster classification on the designated patent document data group based on the part of speech information (2) table and generating cluster information indicating the result of cluster classification.

  The cluster classification process in this embodiment will be described.

  In the cluster classification process, the cluster analysis unit 104 has a morpheme (hereinafter referred to as “first classification”) in which the part of speech in the part of speech information (2) table is a noun, a symbol, an unknown word, or an adjective (hereinafter referred to as “first classification”). And TF (Term Frequency) value and DF (Document Frequency) value in each designated patent document data to calculate “TF value × IDF (Inverse Document Frequency) value” (hereinafter referred to as “subject word candidate target data”). TF / IDF value ") and issue document vector information generation processing for generating assignment document vector information indicating the calculation result, and each designated patent document data having the TF / IDF value of the assignment document vector information as elements. And a similarity calculation process for calculating a similarity between designated patent document data by calculating a cosine value between the designated patent document data using the document vector of each designated patent document data. .

  In addition, a process of clustering the document vectors of the combination of designated patent document data with the maximum similarity as belonging to one group, and the similarity between the cluster after clustering and another document vector or another cluster Cluster generation processing that recalculates the degree using the longest distance method, repeats the above clustering process, sequentially generates clusters, and generates cluster information that associates the specified patent document data belonging to each cluster with the cluster Including.

  Here, the recalculation of the similarity using the longest distance method means, for example, that a document vector of patent document data that does not belong to a cluster (hereinafter referred to as “unclustered document vector”) and a certain cluster. When recalculating the similarity, calculate the similarity between each document vector and unclustered document vector of each patent document data belonging to the cluster, and recalculate the minimum similarity among the calculated similarities To derive as a result. The case of recalculating the similarity between clusters is the same as described above, and the minimum similarity among the calculated similarity between the document vectors of each cluster is derived as a recalculation result between the clusters.

  In this embodiment, since the longest distance method is used, cluster generation is repeated if the maximum similarity value is greater than 0. When the maximum value is 0, cluster generation is terminated and the cluster generation is terminated. Information shall be generated. When a method other than the longest distance method is used, cluster generation is repeated until the maximum value of similarity reaches a predetermined threshold, and generation of the cluster is terminated when the maximum value reaches the predetermined threshold. Cluster information is generated.

  Next, the solution word extraction unit 105 will be described.

  The solution word extraction unit 105 is based on the solution word extraction processing function for extracting the solution word in the designated patent document data group based on the document-specific part-of-speech information (1) table and the cluster information generated by the cluster analysis unit 104. And a matrix element generation processing function for calculating a GF (Global Frequency) value in each cluster of each solution word and generating a matrix element table in which each cluster is associated with the GF value of each solution word.

  Here, the solution word extraction process 105 is performed by the solution word extraction unit 105 for each morpheme whose part of speech in the document-specific part of speech information (1) table corresponds to a predetermined case particle, before the morpheme of the case particle. (Hereinafter, referred to as “forward morpheme”), the morpheme between other predetermined case particles (hereinafter, referred to as “forward morpheme for each case particle”), and the part of speech is classified into the first category. Forward morphemes (hereinafter referred to as “solution candidate morpheme data”) belonging to, and for each solution word candidate morpheme data, a GF value is calculated as an appearance frequency in the designated patent document data group, and the GF values are descended in descending order. Includes calculation processing for determining the rank of each morpheme and solution word list generation processing for extracting a predetermined number of solution word candidate morpheme data in descending order of GF values as solution words in the designated patent document data group and generating a solution word list .

  Note that the predetermined case particles in the present embodiment are the case particles “ha” and “ga”. Since it is considered that a character string that is the subject or purpose is described before these case particles, in this embodiment, attention is paid to the case particles “O” and “GA” in the solution information of each patent document data. Then, words that are important or characteristic in the solution information of the patent document data are extracted from the character strings written in front of these case particles.

  Next, the feature word extraction unit 105 will be described.

  The feature word extraction unit 105 has a feature word extraction processing function that extracts feature words in the designated patent document data group based on the document-specific part-of-speech information (1) table.

  Hereinafter, the feature word extraction process will be described.

  In the feature word extraction process, the feature word extraction unit 105 detects the part-of-speech in the document-specific part-of-speech information (1) table for each predetermined case particle, immediately before the case particle among the front morphemes for each case particle. Forward morpheme combining processing for generating combined morpheme (1) by combining consecutive forward morphemes from the front morpheme until the detection of forward morpheme whose part of speech is other than the first category is detected, and combining in the specified patent document data group A calculation process for calculating the GF value and the DF value of the morpheme (1) and calculating the GF value × IDF value (hereinafter referred to as “GF / IDF value”), and the combined morpheme (1) including the solution word Among them, the top predetermined number of combined morphemes (1) are extracted as feature words in descending order of GF / IDF values, and feature word extraction information for each solution word is generated by associating the extracted feature words with the corresponding solution words. And feature word extraction information generation processing.

  Next, the task word extraction unit 107 will be described.

  The task word extraction unit 107 has a task word extraction processing function for extracting a task word of the designated patent document data group based on the part-of-speech information (2) table.

  In the task word extraction process, the task word extraction unit 107 extracts task word candidate target data in the part-of-speech information (2) table, and based on the document-specific part-of-speech information (2) table including the task word candidate target data, Processing to generate the combined morpheme (2), and the calculation processing to read the cluster information generated by the cluster analysis unit 104 and calculate the GF / IDF value of each combined morpheme (2) in each cluster indicated by the cluster information Then, the top predetermined number of combined morphemes (2) are determined as task words in descending order of GF / IDF values for each cluster, and a task word extraction list in which information indicating each cluster is associated with the determined task word is generated. Task word extraction list generation processing.

  Here, the morpheme combining process refers to a process of referring to the identification numbers in the document-specific part-of-speech information (2) table for each morpheme data of the task word candidate target data and combining the morphemes in which the identification numbers are continuous. It is.

  Based on the cluster information received from the cluster analysis unit 104 and the assignment document vector information, the map generation unit calculates a centroid vector of the cluster from the document vector of each patent document data belonging to each cluster, and An intercluster similarity calculation function that calculates the similarity between each cluster by calculating the cosine value, and an output that defines a row on the matrix that outputs each task word indicating each cluster based on the similarity between each cluster An output order determination processing function for generating rank information.

  Further, the map generation unit 108 calculates a GF value in each cluster of each solution word based on the solution word extraction information and the cluster information, and calculates an inner product value of vectors for each solution word having the GF value as an element. Output rank information that defines a column on a matrix for outputting each solution word and a feature word corresponding to each solution word based on a function for calculating similarity between solution words and an inner product value between each solution word And an output order determination processing function to be generated.

Here, the output order determination process in the present embodiment will be described.
7 and 8 are diagrams for explaining the output order determination process, and show an example of determining the output order of clusters. Since the output order determination process of the solution word is the same, a cluster will be described as an example.

  It is assumed that the output order information in the present embodiment is managed by a linked list structure, and information on a to g indicating each cluster is stored in advance at a predetermined address. Each node in the linked list holds link information indicating a link to the address of the next node, and the order of the nodes indicated by the link information indicates the output order of each cluster.

  The linked list generation process for determining the output order will be described below with reference to FIGS.

  FIG. 6A shows similarity information 310 indicating the result of calculating the similarities of the clusters (a to g) 311 and 312, and the numerical value in the cell of the similarity information 310 is the similarity in this example. The descending order of degrees is shown, and the similarity of the cluster (b, f) indicated by the cell 313 is the maximum.

  FIG. 6B shows the result of associating the similarity ranking 321 with the similarity combination 322 based on the similarity information 310 of FIG. 5A and sequentially extracting the cluster combinations in descending order of similarity. Is shown.

  Hereinafter, the process of determining the output order of clusters will be described with reference to FIG.

  (1) The node b332 and the node f342 of the combination (b, f) having the maximum similarity are set as reference nodes (hereinafter referred to as “reference node b” and “reference node f”), and the HEAD pointer 331 and the HEAD pointer 341 is linked to each reference node to generate each linked list (hereinafter referred to as “first reference linked list 330” and “second reference linked list 340”). When generating each reference linked list, the insertion direction of the node in each reference linked list is inserted between the HEAD side, that is, between the HEAD pointer and the reference node, or added to the TAIL side, that is, the end node in each reference linked list. Set either one of these. However, the node insertion directions of the first reference linked list 330 and the second reference linked list 340 are set to different directions.

  In this example, the node insertion direction of the first reference linked list 330 is set to the HEAD side, and the node insertion direction of the second reference linked list 340 is set to the TAIL side.

  (2) Subsequently, it is determined whether or not each node of the combination (b, d) having the second similarity rank is linked to any reference linked list, and if only one node is linked, The other unconnected node is linked according to the node insertion direction of the reference linked list.

  In this example, since the node insertion direction of the first reference linked list 330 to which the reference node b332 is linked is the HEAD side, the node d333 is inserted between the HEAD pointer 331 and the reference node b332, and the link is changed.

  (3) Next, the same processing as in the above (2) is performed for each node of the combination (f, g) having the third similarity ranking.

  In this example, since the node insertion direction of the second reference linked list 340 to which the reference node f342 is linked is the TAIL side, the node g343 is added to the TAIL of the second reference linked list 340.

  (4) Next, for each node of the combination (a, c) having the fourth similarity ranking, whether or not the node a and the node c are connected is determined in the same manner as (2). If both are unconnected, a reference linked list of each reference node is generated using each node as a reference node.

  In this example, a node a is a reference node a351 and a node c is a reference node c361, and a reference linked list (hereinafter referred to as a “third reference linked list 350” and a “fourth reference linked list 360”) is generated. To do.

  (5) Subsequently, for each node of the combination (b, e) having the fifth similarity rank, the same determination as in (2) above is performed, and the first reference linked list 330 to which the reference node b332 is linked is determined. Since the node insertion direction is the HEAD side, the node e334 is inserted between the HEAD pointer 331 and the node d333, and the link is changed.

  Next, the processes after (6) will be described with reference to FIG.

  (6) If all the elements have been linked to any of the standard linked lists, the number of nodes linked to each standard linked list is counted, and according to the combination of the standard nodes when the standard linked list is generated, A linked list is generated by linking the reference linked list.

  In this example, the first reference linked list 330 of the reference node b332 and the second reference linked list 340 of the reference node f342 are linked to generate the first linked list 370, and the third reference linked list of the reference node a , The second linked list 380 is generated by linking the fourth reference linked list of the reference node c.

  Note that when generating a linked list by linking two reference linked lists, the reference linked lists are linked so that the reference nodes of each reference linked list are linked to each other, and the links are replaced. Also, if the node insertion direction of the reference linked list with the smaller number of nodes in the linked reference linked list is TAIL, the HEAD pointer of the linked list is replaced with the HEAD pointer at the end node of the linked list, and each node Relink the link. If the node insertion direction of the reference linked list with the smaller number of nodes is HEAD, a HEAD pointer is attached to the first node of the reference linked list.

  In this example, when the first reference linked list and the second reference linked list are linked, a link is established between the reference node b332 and the reference node f342, and the HEAD pointer is set to the last node g343 of the second reference linked list 340 having a small number of nodes. 331 is replaced, and the first linked list 371 is generated by replacing the links of the nodes as shown in FIG.

  Similarly, the third reference linked list and the fourth reference linked list are also linked to generate the second linked list 380 shown in FIG.

  (7) Subsequently, the first linked list 371 and the second linked list 380 are linked to generate a rank determination linked list 390 indicating the output rank of the cluster.

  When generating a linked list, the linked list is linked in descending order of similarity of combinations of reference nodes included in the linked list.

  In this example, when the similarity between the reference node b and the reference node f in the first linked list and the similarity between the reference node a and the reference node c in the second linked list are compared, as shown in FIG. The similarity of the linked list is large. Therefore, the node a351 and the node c361 are added in order from the HEAD pointer 381 of the second linked list 380 to the end node e334 of the first linked list, and the rank determining linked list 390 shown in FIG.

  The map generation unit 108 sends the link information of the linked list generated by the above method to the output control unit 109 as cluster output order information.

  Next, the output control unit 109 will be described.

  Based on the cluster output rank information, the solution word output rank information, and the matrix element table generated by the map generation unit 108, the output control unit 109 executes the task word, the solution word and the feature word, and each solution word. It has a function of generating information such as images and data to be arranged in each matrix and each cell of the problem / solution matrix in which the GF value is expressed on a matrix and sending it to the display unit 4.

<Data>
Data used by the information processing apparatus 1 according to the present embodiment will be described.

  FIG. 2A shows the structure of a part-by-document part-of-speech information (1) table generated by the morpheme analysis unit 103 by performing morpheme analysis processing on each piece of solution information included in each patent document data of the designated patent document data group. And an example of data is shown.

  The document-specific part-of-speech information (1) table is used when the solution word extraction unit 106 and the feature word extraction unit 105 extract the solution word and the feature word of the designated patent document data group, and the patent ID 111, the identification number 112, the morpheme Data 113 and part of speech 114 are stored in association with each other.

  Here, the identification number 112 indicates an identification number for identifying each morpheme data decomposed for each solution information of patent document data by morpheme analysis processing, and the morpheme analysis unit 103 detects morpheme data for each solution information. Numbers are assigned in the order in which they are performed. The morpheme data 113 indicates the morpheme data of the resolution information decomposed by the morpheme analysis process, and the part of speech 114 indicates the part of speech to which each morpheme data belongs.

  FIG. 6B shows the part-of-speech for each document generated by the morphological analysis unit 103 by performing morphological analysis processing on the problem information included in each patent document data of the designated patent document data group, as in FIG. The structure of the information (2) table and data examples are shown.

  Part-of-speech part-of-speech information (2) table 120 is used when cluster analysis unit 104 or task word extraction unit 107 performs cluster classification processing and task word extraction, and includes patent ID 121, identification number 122, morpheme data 123, and The parts of speech 124 are stored in association with each other, and the details are the same as those in the document-specific part of speech information (1) table 110 described above.

  FIG. 5C shows a part-of-speech information (2) table.

  The part-of-speech information (2) table 125 in the figure is obtained by integrating the morpheme data of the above-mentioned document-specific part-of-speech information (2) table 120 by excluding duplicate morpheme data, and is used in the cluster classification process. ID No. 126, morpheme data 127, and part of speech 128 are stored in association with each other.

  The identification number 126 is given again when integrating the morpheme data in the document-specific part-of-speech information (2) table 120, and does not indicate the detection order of the morpheme data in each patent document data. The morpheme data 127 indicates morpheme data excluding duplicate morpheme data in all morpheme data of the document-specific part-of-speech information (2) table 120, and the part-of-speech 128 indicates the part-of-speech to which the morpheme data 127 belongs.

Next, FIG. 4D will be described.
FIG. 6D shows the configuration and data example of assignment document vector information.

  The problem document vector information 130 in FIG. 10 is a result of calculating the TF / IDF value in each designated patent document data of the problem word candidate target data extracted by the cluster analysis unit 104 from the part-of-speech information (2) table 125 described above. This is used when the cluster analysis unit 104 classifies the specified patent document data group into clusters.

  Here, the assignment document vector information 130 associates the identification number 131, the morpheme data 132, and the TF / IDF value 133 for each patent document data, and the identification number 131 is the extracted candidate word candidate. Part-of-speech information of target data (2) Indicates an identification number in the table 125, morpheme data 132 indicates target word candidate target data, and TF / IDF value 133 indicates TF of each target word candidate target data in each designated patent document data -Indicates the IDF value.

Next, FIG. 3 will be described.
FIG. 3A shows a configuration of cluster information and an example of data.

  The cluster information 140 in FIG. 6 shows the result of the cluster analysis unit 104 classifying the designated patent document data group into clusters.

  The cluster information 140 is used when the task word extraction unit 107 extracts task words and the map generation unit 108 performs output rank determination processing for each cluster. The cluster information 140, the patent ID 142, and the number 143 are stored in association with each other. Yes.

  Here, the cluster 141 indicates information indicating each cluster generated at the time of cluster classification, and the patent ID 142 indicates the identification ID of designated patent document data belonging to each cluster. The number of cases 143 indicates the number of designated patent document data belonging to each cluster.

  FIG. 3B shows the structure of the combined morpheme (2) information indicating the calculation result of the GF / IDF value in each cluster of the combined morpheme (2) and each combined morpheme (2) combined by the task word extraction unit 107. An example of data is shown.

  The combined morpheme (2) information 150 is used when the task word extraction unit 107 extracts task words, and is stored in association with the combined morpheme (2) 151 and the GF / IDF value (rank) 152 for each cluster. Yes.

  Here, the combined morpheme (2) 151 indicates the data of the combined morpheme (2), and the GF • IDF value (rank) 152 indicates the GF • IDF value in each cluster of each combined morpheme (2), and in each cluster. The ranking of GF / IDF values is shown.

Next, FIG. 3C will be described.
FIG. 4C shows the configuration and data example of the task word extraction list.

  The task word extraction list 160 in FIG. 8 shows task words extracted for each cluster by the task word extraction unit 107 based on the rank of the GF / IDF values of the combined morpheme (2) information 150 described above.

  Here, the task word extraction list 160 stores clusters 161 and task words (1-5) 162 in association with each other, and the cluster 161 indicates information indicating each cluster, and task words (1-5). Reference numeral 162 denotes a task word in each cluster.

  For example, in the same figure, when the cluster 161 is “task 1”, the task word 1 “decrease”, the task word 2 “density unevenness”, the task word 3 “gasiness, etc.” The task word 4 “long term” and the task word 5 “possible” are extracted.

Next, FIG. 4 will be described.
FIG. 4A shows a configuration and data example of the solution word candidate morpheme information.

  The solution word candidate morpheme information 170 shown in the figure is based on the solution word candidate morpheme data extracted from the document-specific part-of-speech information (1) table 110 when the solution word extraction unit 106 extracts solution words and feature words. Generated.

  Here, the solution word candidate morpheme information 170 stores the morpheme 171, the GF value 172, and the rank 173 in association with each other, the morpheme 171 indicates the morpheme data of the solution word candidate, and the GF value 172 indicates each solution. The calculation result of the GF value in the designated patent document data group of the word candidate morpheme is shown, and the rank 173 indicates the rank when the GF values 172 are arranged in descending order.

  FIG. 5B shows the configuration of the solution word list and data examples.

  The solution word list 180 in the figure stores solution words in the designated patent data group extracted by the solution word extraction unit 106 based on the solution word candidate morpheme information 170.

  FIG. 6C shows the configuration and data example of combined morpheme (1) information. The combined morpheme (1) information is used when the feature word extraction unit 105 extracts feature words.

  The combined morpheme (1) information 190 in the figure stores the combined morpheme (1) 191, the GF / IDF value 192, and the morpheme 193 in association with each other.

  Here, the combined morpheme (1) 191 indicates each piece of data of the combined morpheme (1) based on the morpheme data 113 of the document-specific part-of-speech information (1) table 110, and the GF / IDF value 192 is The GF / IDF value in the specified patent document data group of each combined morpheme (1) is shown, and the morpheme 193 shows information on the morpheme constituting each combined morpheme (1).

  Note that each morpheme of the morpheme 193 is forward morpheme data extracted by the solution word extraction unit 106 for each case particle, and when each morpheme data of the morpheme 193 is included in the combined morpheme (1), Is represented by “1”, and “0” when not included. For example, when the combined morpheme (1) 191 in the figure is “magnetic toner”, the morpheme data “toner” is included, so “1” is stored in the “toner” column of the morpheme 193.

Next, FIG. 5 will be described.
FIG. 5A shows the configuration and data example of the solution-specific combined morpheme (1) information.

  In the combined morpheme (1) information 210 for each solution word in the figure, the feature word extraction unit 105 extracts the combined morpheme (1) including the solution word based on the above-described combined morpheme (1) information 190 and the solution word list 180. Shows the results.

  Here, the solution-specific combined morpheme (1) information 210 stores the solution word 211 and the connection morpheme (1) 212 in association with each other, and the solution word 211 is stored in the solution word list 180. Each solution word is shown, and a combined morpheme (1) 212 indicates a combined morpheme (1) including the solution word 211.

  FIG. 5B shows the configuration and data example of the solution word-specific feature word extraction information.

  The feature word extraction information 220 for each solution word in the figure is a result of the feature word extraction unit 105 extracting feature words based on the above-mentioned solution word combination morpheme (1) information 210 and the combination morpheme (1) information 190. .

  The solution word-specific feature word extraction information 220 stores the solution word 221 and the feature word (1-5) 222 in association with each other. The solution word 221 indicates the solution word in the solution word list 180, and the feature word 222. The combined morpheme (1) information 210 of each solution word of the solution morpheme (1) information 210, the rank of the GF · IDF value 192 of the bond morpheme (1) information 190 to the top five in each solution word Each coupled morpheme (1) is shown.

Next, FIG. 3C will be described.
FIG. 3C shows the configuration and data example of the similarity information between clusters.

  The inter-cluster similarity information 230 in FIG. 6 shows the result of the similarity between the clusters calculated by the map generation unit 108 based on the cluster information 140 and the assignment document vector information 130. The inter-cluster similarity information 230 is used when the map generation unit 108 determines the output order of each cluster.

  The inter-cluster similarity information 230 is obtained by associating information 231 and 232 indicating each cluster, and the value of the cell indicated by the information 231 and 232 indicating each cluster is the similarity between the clusters indicated by the clusters. Is shown.

Next, FIG. 6 will be described.
FIG. 6A shows the configuration and data example of the matrix element table.

  The matrix element table 240 in FIG. 5 shows the result of the map generation unit 108 calculating the GF value in each cluster for each solution word, and is used when determining the column to output the solution word.

  The matrix element table 240 stores the solution word 241 and each cluster (task 1 to task n) 242 in association with each other.

  Here, the solution word 241 indicates a solution word stored in the solution word list 180, and the cluster value 242 indicates the GF value of each solution word in each cluster indicated by the tasks 1 to n. .

  FIG. 2B shows the configuration and data example of the similarity information between solution words.

  The solution similarity information 250 in the figure is used when the map generation unit 108 determines a column to output each solution word.

  In the solution word similarity information 250, the solution word 251 and the solution word 252 indicate each solution word in the solution word list, and the cells indicated by the solution words 251 and 252 indicate the similarity value between the solution words. Yes.

<Operation>
Hereinafter, the operation of the information processing apparatus 1 according to the present embodiment will be described.

  10 and 11 show an operation flow of the information processing apparatus 1 according to the present embodiment, which will be described with reference to FIG.

  When the input receiving unit 101 of the information processing apparatus 1 receives an input designating information indicating a technical field (hereinafter referred to as “technical field information”) from the user via the input unit 3, the input receiving unit 101 receives the input from the user. Technical field information is transmitted (step S11).

  The data acquisition unit 102 acquires M designated patent document data groups corresponding to the technical field indicated by the technical field information received from the input reception unit 101 in step S11 from the storage unit 2, and stores the acquired designated patent document data group. Each subject information in each designated patent document data and each claim scope data are extracted and sent to the morpheme analysis unit 103 (step S12).

  The morpheme analysis unit 103 performs morpheme analysis processing on each piece of subject information of each designated patent document data received from the data acquisition unit 102 in step S12, and the document-specific part-of-speech information (2) table 120 for each designated patent document data (FIG. 2). (b)) is generated, the overlapping morpheme data in the document-specific part-of-speech information (2) table is integrated, and the part-of-speech information (2) table 125 (FIG. 2 (c)) in the specified patent data group is generated. 2 (step S13).

  Subsequently, the morpheme analysis unit 103 performs an analysis object extraction process to be a morpheme analysis target for each solution information of each claim range data of each designated patent document data received from the data acquisition unit 102 in step S12 (step S12). S14).

Here, the analysis target extraction process will be described with reference to FIG.
FIG. 12A is a diagram exemplifying solution information described as [Claim 1] 20 in the claims data.

  The morphological analysis unit 103 is characterized in that the solution information is “indicated by the underlined portion 22” (hereinafter referred to as “first character string”) in the “underlined portion 21” of the drawing. "(Hereinafter referred to as" second character string "), the character string between the first character string and the second character string is extracted as an analysis target. .

  FIG. 5B shows a flow of processing for extracting the analysis target, and the analysis target extraction operation will be described with reference to FIG.

  The morpheme analysis unit 103 extracts each piece of solution information from each claim range data of each designated patent document data (step S401), and determines whether each solution information matches a predetermined format (step S402).

  In step S402, when the morpheme analysis unit 103 determines that the solution information matches the predetermined format (step S402: Y), the morpheme analysis unit 103 sets the predetermined data of the first character string and the second character string in the solution information. Each character string in the range between the two is extracted as analysis target data (step S403).

  In step S402, when the morpheme analysis unit 103 determines that the solution information does not match the predetermined format (step S402: N), the morpheme analysis unit 103 extracts all character strings in the solution information as analysis target data. (Step S404).

  Returning to FIG. 10, the morpheme analysis unit 103 performs morpheme analysis processing on the data of the analysis target portion of the solution information extracted in step S14, and for each patent document data, the document-specific part-of-speech information (1) table 110 (FIG. )) Is generated and stored in the storage unit 2 (step S15).

  The cluster analysis unit 104 performs clustering processing for classifying the designated patent document data group into a plurality of clusters based on the part-of-speech information (2) table 125 (step S16).

Hereinafter, the clustering process will be described with reference to FIG.
The cluster analysis unit 104 reads the part-of-speech information (2) table 125 from the storage unit 2 (step S501). Of the morpheme data stored in the morpheme data 123 of the part-of-speech information (2) table 125, the part of speech 124 is the first. Morphological data belonging to the classification is extracted as task word candidate target data (step S502).

  The cluster analysis unit 104 calculates the TF / IDF value in each designated patent document data for each subject word candidate target data extracted in step S502, and generates the subject document vector information 130 (FIG. 2 (d)). In step S503, a document vector for each designated patent document data having the TF / IDF value of the assignment document vector information 130 as an element is generated (step S504).

  Subsequently, the cluster analysis unit 104 calculates a cosine value between the document vectors by using each document vector for each designated patent document data generated in step S504 to obtain a similarity between the designated patent document data (step S505). ).

  The cluster analysis unit 104 sets the counter to 1 (step S506), and generates a cluster by integrating the document vectors of the designated patent document data having the maximum similarity among the similarities obtained in step S505 (step S506). Step S507).

  The cluster analysis unit 104 determines whether or not the counter value is smaller than the designated patent document data number M-1 (step S508). If the counter value is smaller than M-1 (step S508: Y), the cluster analysis unit 104 is generated in step S507. The similarity between each document vector of the cluster and the remaining document vectors is recalculated using the longest distance method (step S509).

  The cluster analysis unit 104 determines whether or not the maximum value of each similarity recalculated in step S509 is greater than 0 (step S510), and if the maximum value of similarity is greater than 0 (step S510: Y). Then, 1 is added to the counter value (step S511), and the processing from step S507 onward is repeated.

  If the cluster analysis unit 104 makes a negative determination in step S510, that is, if the maximum similarity is 0 (step S510: N), each cluster generated as a result of document vector integration processing is displayed. The number of designated patent document data belonging to the cluster is counted, and cluster information 140 (FIG. 3A) in which information indicating the cluster is associated with the designated patent document data belonging to the cluster is generated, and the storage unit 2 (step S512).

  Returning to FIG. 10, the task word extraction unit 107 performs a task word extraction process of extracting task words indicating each cluster generated in step S <b> 16 based on the document-specific part-of-speech information (2) table 120 and the cluster information 140. This is performed (step S17).

Details of the task word extraction process will be described below with reference to FIG.
14, the task word extraction unit 107 reads the part of speech information (2) table 125 and the document-specific part of speech information (2) table 120 from the storage unit 2 (step S601), and the part of speech information (2) table 125 reads the part of speech. Extracts task word candidate target data belonging to the first category (step S602), performs morpheme combining processing using the document-specific part-of-speech information (2) table 120 including each extracted subject word candidate target data, 2) is generated (step S603).

Here, the morpheme combining process will be described with reference to FIG.
In step S310 of FIG. 15, the task word extraction unit 107, in the document-specific part-of-speech information (2) table 120, the morpheme data corresponding to the patent ID including the morpheme data matching each task word candidate target data extracted in step S601. The identification number of each morpheme data is read with reference to the identification number.

  The task word extraction unit 107 generates a combined morpheme (2) by combining the continuous morpheme data as long as the identification numbers continue based on the identification numbers read in step S310 (step S311).

  The task word extraction unit 107 determines whether or not the combining process has been completed for all the extracted morpheme data of the candidate word candidates (step S312), and when the combining process of all the morpheme data has been completed (step S312). : Y), the morpheme combining process in step S311 is ended. If not (step S312: N), the processes in and after step S311 are repeated.

  Returning to FIG. 14, the task word extraction unit 107 reads the cluster information 140 from the storage unit 2 (step S604), and for each combined morpheme (2) generated in step S311 described above, patents belonging to each cluster of the cluster information 140 The GF / IDF value in the document data group is calculated, and the rank is derived in descending order of the GF / IDF value to generate the combined morpheme (2) information 150 (FIG. 3B) (step S605).

  Subsequently, the task word extraction unit 107 refers to the GF / IDF value 152 for each cluster in the combined morpheme (2) information 150, and determines each combined morpheme (2) having the top five GF / IDF values for each cluster. A task word extraction list 160 (FIG. 3C) is generated and stored in the storage unit 2 (step S606).

  Returning to FIG. 10, the solution word extraction unit 106 and the feature word extraction unit 105 perform an extraction process for extracting the solution word and the feature word in the designated patent document data group based on the document-specific part-of-speech information (1) table 110 ( Step S18).

The solution word and feature word extraction processing will be described below with reference to FIG.
In step S701 in FIG. 16, the solution word extraction unit 106 reads out the document-specific part-of-speech information (1) table 110 (FIG. 2A) from the storage unit 2, and the part-of-speech 114 in the document-specific part-of-speech information (1) table 110 The forward morpheme data is extracted for each morpheme data of a predetermined case particle, and the solution candidate morpheme data whose part of speech belongs to the first classification is extracted from the extracted front morpheme data for each extracted case particle (step S702).

  Here, for example, in the case of the record 115 in which a predetermined case particle “O” is stored in the document-specific part-of-speech information (1) table 110, the front morpheme data of this case particle is expressed by “magnetic” “brush” ”. Among these forward morpheme data, “magnetic”, “brush”, and “toner” belonging to the first classification (noun, unknown word, symbol, adjective) are extracted as solution candidate morpheme data. .

  Subsequent to step S702, the solution word extraction unit 106 calculates the GF value in the designated patent document data group for each solution word candidate morpheme data, derives the rank in descending order of the GF value, and sets the solution word candidate target information 170 ( FIG. 4A) is generated (step S703).

  The solution word extraction unit 106 determines, as solution words, morphemes in which the rank 173 of the solution word candidate target information 170 generated in step S703 reaches a predetermined rank, and generates a solution word list 180 (FIG. 4B) (steps). S704).

  The solution word extraction unit 106 extracts the forward morpheme for each predetermined case particle from the document-specific part-of-speech information (1) table 110 as in step S702 (step S705). Part-of-speech part-of-speech information (1) In the table 110, in order from the front morpheme detected immediately before the case particle, each front morpheme is combined in the detection order until front morpheme data other than the first classification appears. The forward morpheme combining process for generating (1) is performed, and the combined morpheme (1) is stored in the combined morpheme (1) information 190 (FIG. 4C). Further, when each combined morpheme (1) includes the morpheme extracted in step S705, the solution word extraction unit 106 stores “1” in the morpheme 193 of the combined morpheme (1) information 190 (step S706). ).

  Here, for example, in the case of the record 115 in which the predetermined case particle “O” is stored in the part of speech 114 of the document-specific part of speech information (1) table 110, the case particle “O” is preceded by the same as step S 702 described above. “Toner” is extracted by “magnetic” “brush” as the morpheme, and “magnetic” “brush” “toner” is specified as the front morpheme whose part of speech belongs to the first classification. The identification numbers in the document-specific part-of-speech information (1) table 110 of these identified front morphemes are “magnetism = 1”, “brush = 2”, “toner = 4”. When the front morphemes having consecutive identification numbers are combined in order from “toner” which is a morpheme, “toner” is generated as a combined morpheme (1). Also, “1” is stored in the morpheme 193 “toner” as the constituent morpheme information of the combined morpheme (1) information 190 (FIG. 4C) “toner”.

  Subsequent to step S706, the solution word extraction unit 106 calculates the GF • IDF value in the designated patent document data group for each combined morpheme (1) generated in step S706, and the GF • ID of the combined morpheme (1) information 190 is calculated. Stored in the IDF value 192 (step S707).

  The feature word extraction unit 105 performs a feature word determination process for determining a feature word corresponding to each solution word based on the combined morpheme (1) information 190 generated in steps S706 and S707 (step S708).

Here, the feature word determination process will be described with reference to FIG.
In step S709 of FIG. 17, the feature word extraction unit 105 refers to the combined morpheme (1) 191 of the combined morpheme (1) information 190, and among the morphemes of the morpheme 193, the solution word stored in the solution word list 180 The combined morpheme (1) in which “1” is stored in the matching morpheme 193 is extracted, and the solution-specific combined morpheme (1) information 210 (FIG. 5A) is generated.

  The feature word extraction unit 105 refers to the GF / IDF value 192 of the combined morpheme (1) information 190 (FIG. 4C) to resolve the combined morpheme (1) of the solution-specific combined morpheme (1) information 210. For each word 211, the top five combined morphemes (1) 212 in the descending order of GF / IDF values are extracted (step S 710), and the extracted combined morphemes (1) 212 are determined as feature words, and the features by solution word Word extraction information 220 (FIG. 5B) is generated and stored in the storage unit 2 (step S711).

  In step S19 of FIG. 11, the map generation unit 108 generates a matrix word for outputting the task word of each cluster based on the cluster information 140 (FIG. 3A) and the task document vector information 130 (FIG. 2C). An output order determination process for determining a row is performed.

  Details of the cluster output order determination processing will be described with reference to FIG.

  In step S <b> 801 of FIG. 18, the map generation unit 108 reads the cluster information 140 and the assignment document vector information 130 from the storage unit 2.

  The map generation unit 108 refers to the TF / IDF value 133 of each patent document data of the assignment document vector information 130, and uses the patent ID of the assignment document vector information 130 as an element of each document vector for each patent ID 142 of the cluster information 140. The TF / IDF value 133 is extracted. Subsequently, the map generation unit 108 calculates a centroid vector for each cluster using the extracted document vector group for each cluster (step S802).

  The map generation unit 108 calculates the cosine value between the centroid vectors using the centroid vector for each cluster calculated in step S802 to obtain the similarity between the clusters, and the intercluster similarity information 230 (FIG. 5C). ) Is generated (step S803).

  Subsequently, the map generation unit 108 sequentially extracts combinations of clusters in descending order of similarity in the inter-cluster similarity information 230 (step S804).

  The map generation unit 108 generates a linked list that defines the output rank information of the clusters for each combination of clusters sequentially extracted in step S804 (step S805).

The linked list generation process will be described below with reference to FIG.
In step S31 of FIG. 19, for each combination of clusters sequentially extracted in step S804, the map generation unit 108 determines whether each cluster (hereinafter referred to as “element”) included in the combination is unconnected. Judge whether or not.

  When the map generation unit 108 determines that all of them are unconnected in step S31 (step S31: Y), it generates a reference linked list using each element of the combination as a reference node, and HEAD or TAIL in each reference linked list Is determined as the node insertion direction (step S32).

  In addition, when the map generation unit 108 determines that none of them is unconnected in step S31 (step S31: N), the map generation unit 108 has only one element of the combination has been connected to the reference linked list. It is determined whether or not there is (step S33).

  If the map generation unit 108 makes a positive determination in step S33 (step S33: Y), the unconnected nodes of the combination are inserted according to the node insertion direction in the reference linked list including the connected nodes of the combination ( Step S34).

  If the map generation unit 108 makes a negative determination in step S33 (step S33: N), that is, if any element of the combination is connected, all nodes indicating all clusters are displayed. Is already connected to the reference link list (step S35).

  If the map generation unit 108 determines in step S35 that all cluster elements have been linked (step S35: Y), the number of nodes in each reference linked list is counted and stored in the memory (step S36).

  The map generation unit 108 links the reference nodes when the reference linked lists are generated in step S32, and links the reference linked lists to generate a linked list (step S37).

  Subsequently, the map generation unit 108 performs linked list rank determination processing for determining the output rank of each linked list generated in step S37 (step S38).

  Details of the linked list order determination process will be described below with reference to FIG.

  In step S41 in FIG. 20, the map generation unit 108 reads out from the memory the number of nodes in each reference linked list counted in step S36 in FIG. 19, and in each linked list, the number of nodes in the reference linked list constituting each linked list. A reference linked list having a smaller value is specified (step S42).

  The map generation unit 108 determines whether or not the node insertion direction of the reference linked list identified in step S42 is TAIL (step S43). If it is determined that the node is on the TAIL side (step S43: Y), the reference linked list The HEAD pointer of the linked list including the list is attached to the end node of the linked list, and the link is changed (step S44).

  If the map generation unit 108 determines in step S43 that it is not the TAIL side (step S43: N), that is, if the node insertion direction of the reference linked list identified in step S42 is HEAD, the reference A HEAD pointer is attached to the first node of the linked list, and the link is changed (step S45).

  Subsequently, the map generation unit 108 connects in order from the first node of each linked list in descending order of similarity between the reference nodes of each linked list, and the order of the nodes indicated by each linked information after the linked list is linked. Are determined as the output order of the task words of each cluster indicated by each node (step S46).

  Returning to FIG. 11, following step S19, the map generation unit 108 assigns each solution word to a matrix based on the solution word list 180 (FIG. 4B) and the cluster information 140 (FIG. 3A). An output order determination process for outputting to a column is performed (step S20).

  The solution word output order determination process will be described below with reference to FIG.

  In step S901 of FIG. 21, the map generation unit 108 reads the solution word list 180 and the cluster information 140 from the storage unit 2, refers to the patent ID 142 of the cluster information 140 and the solution word of the solution word list 180, and patents for each cluster. The matrix element table 240 (FIG. 6A) is generated by calculating the GF value of each solution word in the document data group (step S902).

  The map generation unit 108 generates a vector for each solution word having each GF value of the matrix element table 240 generated in step S902 as an element (step S903), calculates an inner product value between the generated document vectors, The similarity between the solution words is obtained, and the solution similarity information 250 (FIG. 6B) is generated (step S904).

  Subsequently, the map generation unit 108 sequentially extracts combinations of solution words in descending order of similarity of the solution word similarity information 250 generated in step S904 (step S905), and each solution word sequentially extracted in step S905. For the combination, a linked list generating process for defining output order information of the solution word is performed (step S906).

  The linked list generation process for the solution word is the same as that in step S805 described above, and a description thereof will be omitted.

  The map generation unit 108 determines the order of each node indicated by the link information of the linked list generated in step S906 as the output order information of the solution word indicated by each node (step S907).

  Returning to FIG. 11, following step S20, the map generation unit 108 includes the feature word extraction information 220 (FIG. 5B), the task word extraction list 160 (FIG. 3C), and cluster information. 140 (FIG. 3 (a)), output order information of clusters and solution words, and a matrix / element table 240 (FIG. 6 (a)), a problem / solution matrix output for generating and outputting a problem / solution matrix Processing is performed (step S21).

  Hereinafter, the problem / solution matrix output processing will be described with reference to FIG.

  The output control unit 109 reads the cluster information 140, the solution word-specific task word extraction list 160, and the cluster output rank information from the storage unit 2 (step S211), and determines the task word indicating the cluster according to the cluster output rank information. Data of a task word indicating a cluster is output to each row of a predetermined area on the matrix to be output (step S212).

  Subsequently, the output control unit 109 reads the feature word extraction information 220 for each solution word and the output order information of the solution word from the storage unit 2 (Step S213), and each solution word and each solution according to the output word information of the solution word. Data of each solution word and feature word is output to each column of a predetermined area on the matrix for outputting the feature word corresponding to the word (step S214).

  Subsequent to step S214, the output control unit 109 reads the matrix element table 240 from the storage unit 2 (step S215), and the GF of the cluster and solution word combination indicated by each row and each column constituting each cell on the matrix. The value is read from the matrix element table 240 and output to each cell (step S216).

  Here, FIG. 9 shows an output example of the problem / solution matrix output by the output control unit 109 after performing the processes in steps S213 to S216.

  In the problem / solution matrix 260 shown in the figure, the predetermined area for outputting the data of each cluster 161 and the problem word 162 in the above-described problem word extraction list 160 is a portion indicated by a broken-line rectangle 262 in the figure, and the rectangle 262 The information 265 indicating the cluster and the task word 266 corresponding to the cluster 265 are output in each row of. The predetermined area for outputting the data of the solution word 221 and the feature word 222 of the solution word-specific feature word extraction information 220 described above is a portion indicated by a broken-line rectangle 261 in FIG. The feature word 263 is output. Further, each data of the GF value 242 of the matrix element table 240 described above is output to each cell indicated by a row and a column on the matrix 260 corresponding to the cluster 242 and the solution word 241 corresponding to each data.

<Discussion>
In the problem / solution matrix in the above-described embodiment, in the cluster and solution word output rank information indicated in the above-described linked list, the reference link having the smaller number of nodes among the two reference linked lists included in the linked list. The end or top node of the list is placed above or to the left of the matrix 260. This is because the larger the GF value in the solution word cluster that is an element of the matrix, the greater the cell output in the upper left cell of the matrix cell region.

  For example, when changing the color of each cell according to the GF value and displaying it, the user determines which solution or feature word of the designated patent document data group is in the designated patent document data group by the color of each cell of the matrix. It can be confirmed at a glance whether the data is frequently used in the similar patent document data group. Conversely, solution words and feature words that are not frequently used in the designated patent document data group can be confirmed at a glance, and the solution words and feature words are characteristic in the designated patent document data group. It can be easily determined whether or not there is.

[Embodiment 2]

<Overview>
The first embodiment described above is similar in the technical field by classifying clusters based on each problem information included in a patent document data group belonging to a certain technical field and representing the characteristics of the patent document data group for each cluster in a matrix. This makes it possible for the user to easily understand the main technology and characteristic technology for each patent document data group.

  In addition to the first embodiment described above, the information processing apparatus according to the present invention allows the user to easily understand the technical features of each patent document data, and at the same time, features the other patent document data. It can be easily compared.

  In the present embodiment, in the problem / solution matrix shown in the first embodiment described above, designation of a certain cluster is accepted from the user, feature words for each patent document data belonging to the cluster are extracted, and each cluster in the cluster is extracted. The directivity of the technical features of the patent document data is output so that the user can easily grasp it.

<Configuration>
The information processing apparatus according to the present embodiment uses the information processing apparatus according to the first embodiment, and the following description is about parts different from the functions of the components described in the first embodiment.

  In the present embodiment, the cluster information 140 (FIG. 3A) and the document-specific part-of-speech information (1) table 110 (FIG. 2A) generated in the first embodiment are stored in the storage unit 2. It shall be.

  The input receiving unit 101 designates a cluster on the above-described problem / solution matrix via the input unit 3 and receives an individual feature analysis instruction for analyzing the characteristics of each patent document data belonging to the cluster from the user. It has a function of sending to the data acquisition unit 102 information indicating a cluster related to the analysis instruction (hereinafter referred to as “analysis target cluster information”) and instruction reception information indicating that the individual feature analysis instruction has been received.

  The data acquisition unit 102 receives the analysis target cluster information and the instruction reception information from the input reception unit 101, refers to the above-described patent ID 142 of the cluster information 140 (FIG. 3A), and determines the cluster indicated by the analysis target cluster information. Each patent document data has a function of acquiring from the storage unit 2.

  The morpheme analysis unit 103 performs the above-described analysis target extraction processing for each claim document data of each patent document data indicated by the patent ID acquired by the data acquisition unit 102, performs morpheme analysis processing, It has a function of generating a document-specific part-of-speech information table.

  In this embodiment, as described above, since the document-specific part-of-speech information (1) table 110 generated in the first embodiment is used, the morphological analysis unit 103 uses the data acquisition unit from the document-specific part-of-speech information (1) table 110. The part-of-speech information (1) corresponding to each patent ID received in 103 is extracted. The extracted part-of-speech information (1) is hereinafter referred to as “document-specific part-of-speech information” and is stored in the storage unit 2 as a document-specific part-of-speech information table to be described later.

  The feature word extraction unit 105 refers to the document-specific part-of-speech information table generated by the morpheme analysis unit 103, and in the same way as in the first embodiment, the part-of-speech of the front morpheme for each case particle belongs to the first morpheme ( Hereinafter, the above-described forward morpheme combination process is performed on “first classification forward morpheme”) to generate a combined morpheme. In the present embodiment, the combined morpheme generated by performing the forward morpheme combining process is hereinafter referred to as “feature word 1”.

  The feature word extraction unit 105 further extracts the case particle for each morpheme whose part of speech is a predetermined case particle "" and "" in the document-specific part of speech information extracted by the morpheme analysis unit 103, as in the first embodiment. Among the morphemes detected later (hereinafter referred to as “backward morphemes”), the backward morphemes detected between other case particles (hereinafter referred to as “backward morphemes for each case particle”) are as follows: Each successive back morpheme until a back morpheme other than a part of speech (hereinafter referred to as “second class”) obtained by adding a verb to the part of speech of the first class is detected from the back morpheme detected immediately after the case particle. It has a function of generating a combined morpheme for each case particle by performing a backward morpheme combining process of combining morphemes (hereinafter referred to as “second classification backward morpheme”) in the order of detection. In the present embodiment, the combined morpheme generated by performing the backward morpheme combining process is hereinafter referred to as “feature word 2”.

  Also, the feature word extraction unit 105 is a feature word extraction list for each document in which feature words 1 and forward morphemes constituting the feature words 1 of each patent document data, feature words 2 and backward morphemes constituting the feature words 2 and the like are associated with each other. A document-specific feature word extraction list generation process, a function of calculating a TF value in each patent document data for the first classification forward morpheme of the patent document data group, and a patent document of the document-specific feature word extraction list For each forward morpheme for each data, it has a function of calculating the GF value in the patent document data group and determining the GF ranking of each forward morpheme in descending order of the GF value.

  The map generation unit 108 generates a document vector having each TF value for each patent document data calculated by the feature word generation unit 105 as an element, calculates an inner product value between the document vectors, and similarity between the patent document data A combination of patent document data is sequentially extracted in descending order of the inner product value between the calculated patent document data and the output position on the output area in the display unit 4 of each output information related to each patent document data And a function of performing order determination processing for determining the output order to be shown.

  In the present embodiment, the output information related to the patent document data includes bibliographic information such as applicant information, application number, and the name of the invention included in each patent document data, and feature words of the patent document data. 1 and feature word 2 (hereinafter referred to as “feature word information”).

  The rank determination process is performed by the same method as the rank determination process of the first embodiment.

  The map generation unit 108 determines the output position of each feature word information in the feature word information output area of the patent document data (hereinafter referred to as “feature word output area”) in each output area of each patent document data. Generating document-specific output order information indicating a function for performing feature word output position determination processing, an output order of each patent document data determined by the order determination process, and an output position of feature word information in the patent document data It has a function.

  The output control unit 109 has a function of sending bibliographic information of patent document data, feature word information data, and an image to the output area of the display unit 4 based on the output order information received from the map generation unit 108.

<Data>
Hereinafter, data used in the information processing apparatus 1 according to the present embodiment will be described.

  FIG. 23 shows the configuration and data example of the document-specific part-of-speech information table generated by the morpheme analyzer 103.

  The document-specific part-of-speech information table 30 shown in the figure is used when the feature word extraction unit 105 generates feature word information. The patent ID 31, the identification ID 32, the morpheme 33, and the part of speech 34 are stored in association with each other. Yes.

  The structure of the document-specific part-of-speech information table 30 is the same as that of the document-specific part-of-speech information (1) table 110 described in the first embodiment, and a description thereof will be omitted.

  FIG. 24 shows the configuration and data example of the document-specific feature word extraction list generated by the feature word extraction unit 105.

  The document-specific feature word extraction list 40 in the figure stores patent ID 41, application number 42, identification ID 43, forward morpheme 44, feature word 1 (45), backward morpheme 46, and feature word 2 (47) in association with each other. ing.

  Here, the patent ID 41 indicates the patent ID of the patent document data indicated by the analysis target cluster information related to the individual feature analysis instruction, and the application number 42 indicates the application number included in each patent document data indicated by the patent ID 41. The identification ID 43 indicates an identification ID of a predetermined case particle in the document-specific part-of-speech information table 30 of the patent document data indicated by the patent ID 41.

  The forward morpheme 44 indicates the first classification forward morpheme used when generating the feature word 1 for the case particle indicated by the identification ID 41, and the feature word 1 (45) indicates each forward morpheme indicated by the forward morpheme 44. The combined morpheme generated by combining is shown. The backward morpheme 46 indicates the second classification backward morpheme used when generating the feature word 2 for the case particle indicated by the identification ID 41, and the feature word 2 (47) represents each of the backward morphemes 46. A combined morpheme generated by combining backward morphemes is shown.

  FIG. 25A shows the structure and data example of the TF value table for each document, and the feature word extraction unit 105 uses each patent document for each front morpheme constituting the feature word 1 and the feature word 2 of all patent document data. The result of calculating the TF value in the data is shown.

  The document-specific TF value table 50 shown in the figure is used when the map generation unit 108 determines the output position of the output information related to each patent document data, and stores the front morpheme 51 and the patent ID 52 in association with each other.

  Here, the forward morpheme 51 indicates all the forward morphemes stored in the forward morpheme 44 of the document-specific feature word extraction list 40, and the patent ID 52 is the patent document data of each patent ID of the individual feature analysis target. Shows the TF value of each forward morpheme.

  FIG. 25B shows the structure and data example of the document-specific GF value table. The feature word extraction unit 105 calculates the GF value in the patent document data group subject to individual feature analysis for each forward morpheme for each patent document data. The calculated result is shown.

  The document-specific GF value table 60 in the figure stores a patent ID 61, a front morpheme 62, a GF value 63, and a GF ranking 64 in association with each other.

  Here, the patent ID 61 indicates the patent ID of each patent document data subject to individual feature analysis, and the forward morpheme 62 indicates the forward morpheme of the patent document data of the patent ID 61. The GF value 63 indicates the GF value in the patent document data group subject to individual feature analysis of each forward morpheme 62 corresponding to the patent ID 61, and the GF ranking 64 indicates the GF value 63 for each patent ID 61 in descending order. The order when arranged is shown.

  FIG. 25C shows the structure and data example of the output order information for each document, and the output position and feature word information output position relating to each patent document data determined by the map generation unit 108. Is shown.

  The output order information 70 stores patent ID 71, document output order 72, feature word 1 (73), forward morpheme 74, and feature word output order 75 in association with each other.

  Here, the patent ID 71 indicates the patent ID of each patent document data subject to individual feature analysis, and the document output order 72 indicates each output position of the output information related to the patent document data of each patent ID. It is assumed that each rank indicated by the document output rank corresponds to each output position.

  The feature word 1 (73) indicates the feature word 1 generated for each patent document data indicated by the patent ID 71, and the forward morpheme 74 is one of the forward morphemes constituting the feature word 1 (73). The forward morpheme that is used as a reference when determining the output rank of the feature word 1 is shown. The feature word output rank 75 indicates the GF rank 64 in the document-specific GF value table 60 of each forward morpheme 74, and each rank indicated by the feature word output rank and the feature word output area of each patent document data. Assume that each feature word output position corresponds.

<Operation>
Next, the operation of the information processing apparatus 1 according to this embodiment will be described.

  FIG. 27 shows an operation flow of the information processing apparatus according to the present embodiment, and the operation will be described with reference to FIG.

  In step S110 of the figure, the information processing apparatus 1 performs a feature word extraction process.

  Details of the feature word extraction process will be described below with reference to FIG.

  The input receiving unit 101 receives input of analysis target cluster information indicating a patent document data group that is an individual feature analysis target from the user via the input unit 3, and sends the analysis target cluster information and instruction reception information to the data acquisition unit 102. It is sent out (step S111).

  When the data acquisition unit 102 receives the analysis target cluster information sent by the input reception unit 101 in step S111, the data acquisition unit 102 reads the cluster information 140 from the storage unit 2, reads the patent ID 142 of the cluster indicated by the analysis target cluster information, and reads the read patent. The patent document data of the ID is acquired from the storage unit 2, and the read patent ID 142 is sent to the morphological analysis unit 103 (step S112).

  When the morphological analysis unit 103 receives the instruction reception information and the patent ID 142 from the data acquisition unit 102, the morphological analysis unit 103 reads the document-specific part-of-speech information (1) table 110 from the storage unit 2, and the patent ID 142 from the document-specific part-of-speech information (1) table 110. The part-of-speech information (1) corresponding to is extracted (step S112), and the document-specific part-of-speech information table 30 (FIG. 23) is generated (step S113).

  The feature word extraction unit 105 reads the document-specific part-of-speech information table 30 from the storage unit 2, and for each patent ID 31 of the document-specific part-of-speech information table 30, the front morpheme and the rear morpheme for each morpheme whose part of speech 34 is a predetermined case particle. Extract (step S114).

  Subsequently, the feature word extraction unit 105 performs the forward morpheme combination process and the backward morpheme combination process for each of the front morpheme and the rear morpheme for each case particle of each patent ID extracted in step S114, and the feature word for each patent ID. 1 and feature word 2 are generated (step S115), and the feature word 1 and feature word 2 generated for each patent ID are associated with the front morpheme and the rear morpheme that constitute each of the feature word 1 and feature word 2. Another feature word extraction list 40 (FIG. 24) is generated (step S116).

  Here, the processing of step S114 and step S115 will be described taking the patent ID “patent 2 of identification ID“ 30 ”of patent 2” in the document-specific part-of-speech information table 30 as an example.

  The feature word extraction unit 105 pays attention to the identification ID “30” corresponding to “a predetermined case particle” in the document-specific part-of-speech information table 30.

  The feature word extraction unit 105 extracts each front morpheme with the identification ID “30” between the identification IDs “30” to “18” (step S114), and the identification ID “29” immediately before the identification ID “30”. In order from the front morpheme, the part of speech 34 is the morpheme of the part of speech “participant” other than the first class, the identification ID “26” of the same figure, each of the front morpheme until “of the morpheme” appears, “toner” “remaining” ” The “development residual toner” is generated as the feature word 1 of the identification ID “30” by combining the “development” in the order of the identification ID (step S115).

  Also, the feature word extraction unit 105 extracts each rear morpheme of the identification ID “30” between the identification IDs “30” to “34” (step S114), and the identification ID “immediately after the identification ID“ 30 ”” In order from the rear morpheme of 31 ", the part of speech 34 is combined with the rear morpheme until the appearance of a morpheme other than the second classification," stripping "," take "," to ", in the order of the identification ID, and the feature word 2 of the identification ID" 30 " Is generated (step S115).

  Returning to FIG. 27, the processing after step S120 will be described.

  Subsequent to step S110, the map generation unit 108 reads the document-specific feature word extraction list 40 from the storage unit 2, and performs output order determination processing for determining the output position of the patent document data (step S120).

  Hereinafter, the process of determining the output order of patent document data will be described with reference to FIG.

  In step S121 of the figure, the map generation unit 108 extracts all the forward morphemes of the forward morpheme 44 in the document-specific feature word extraction list 40, calculates the TF value in each patent document data of the individual feature analysis target, Another TF value table 50 (FIG. 25A) is generated.

  Subsequently, the map generation unit 108 generates a document vector for each patent ID including each TF value for each patent ID in the document-specific TF value table 50 (step S122), and calculates an inner product value between the document vectors. (Step S123).

  The map generation unit 108 sequentially extracts combinations of patent IDs in descending order of the inner product values calculated in step S123 (step S124), and performs a linked list generation process for generating output order information of patent document data (step S125). ).

  Note that the detailed description of the linked list generation process is the same as that in the first embodiment, and thus the description thereof is omitted here.

  The map generation unit 108 determines the order of the nodes indicated by the link information of the linked list generated in step S125 as the output order of the patent document data indicated by each patent ID, and the output order information 70 for each document (FIG. 25 (c)). The determined rank is stored in the document output rank 72 corresponding to each patent ID 71 (step S126).

  Returning to FIG. 27, following step S120, the map generation unit 108 performs an output order determination process for determining an output position in the output area of the feature word information of each patent document data (step S130).

  The details of the feature word information output order determination process will be described below with reference to FIG.

  The map generation unit 108 reads the document-specific feature word extraction list 40 (FIG. 24) from the storage unit 2, and calculates the GF value in the patent document data group of the individual feature analysis target for each front morpheme 44 for each patent ID of the patent ID 41. The GF ranking of the forward morpheme 44 for each patent ID is determined in descending order of the calculated GF value (step S131), and the document-specific GF value table 60 is generated (step S132).

  Subsequently, the map generation unit 108 refers to the forward morpheme 44 in the document-specific feature word extraction list 40 (step S133), and for each feature word 1 (45), the feature word 1 is composed of a plurality of forward morphemes. It is determined whether or not there is (step S134).

  If the map generation unit 108 determines in step S135 that the feature word 1 is composed of a plurality of forward morphemes (step S134: Y), the map generation unit 108 stores each feature word 1 (45) in the document-specific feature word extraction list 40. Among the front morphemes 44, in the document-specific GF value table 60, the front morpheme with the GF rank 64 of the front morpheme of the feature word 1 identifies the highest morpheme and its rank, and the identified forward morpheme and rank are output by document. The forward morpheme 74 corresponding to the feature word 1 in the rank information 70 and the feature word output rank 75 are set (step S135).

  If the map generation unit 108 determines in step S134 that the feature word 1 is not composed of a plurality of forward morphemes (step S134: N), that is, the feature word 1 is composed of one forward morpheme. In the document-specific GF value table 60, the forward morpheme 62 that matches the feature word 1 and the GF rank 64 of the forward morpheme are represented by the forward morpheme corresponding to the feature word 1 of the document-specific output rank information 70. 74 and feature word output ranking 75 are set (step S136).

  Returning to FIG. 27, following step S <b> 130, the output control unit 109 performs output processing of a feature word map representing each output information of the patent document data group of the individual feature analysis target based on the document output order information 70. (Step S140).

  Details of the feature word map output processing will be described below with reference to FIG.

  The output control unit 109 reads out the document-specific output order information 70 (FIG. 25C) from the storage unit 2, and each bibliography of the patent document data for individual feature analysis acquired by the data acquisition unit 102 in step S112 of FIG. Target information is read out (step S141).

  Subsequently, the output control unit 109 refers to the document output order 72 of the document output order information 70, and stores the bibliographic information of each patent document data in the output area for each patent document data corresponding to each document output order. An instruction to output data is sent to the display unit 4 (step S142).

  The output control unit 109 refers to the feature word output order 75 of the document-specific output order information 70, and in the feature word information output area of each patent document data, the patent document data is output to the output position corresponding to each feature word output order. An instruction to output the feature word 1 is sent to the display unit 4 (step S143).

  Subsequent to step S143, the output control unit 109 reads the document-specific feature word extraction list 40 (FIG. 24), and each feature word 2 corresponding to each feature word 1 of each patent document data for which an output instruction has been issued in step S143. Is sent to the display unit 4 at the output position corresponding to the output position of each feature word 1 instructed in step S143 (step S144).

  Here, an output example of the feature word map displayed by the display unit 4 receiving the above-described output instruction from the output control unit 109 is shown in FIG.

  In the feature word map 80 in the figure, each output area indicated by broken-line rectangles 81a to 81c (hereinafter referred to as "output areas 81a, 81b, 81c") indicates an output area for each patent document data. Each of these output areas corresponds to the output order of patent document data, and in this embodiment, output information of patent document data having a higher output order is displayed in order from the upper output area.

  Next, bibliographic information and output positions of feature word information in the output areas 81a, 81b, and 81c will be described using the output area 81a as an example.

  Each output area indicated by a rectangle 82 and a rectangle 84 (hereinafter referred to as “output area 82” and “output area 84”) is an area in which bibliographic information of patent document data to be output to the output area 81a is output. The output area 82 displays the data of the application number of the patent document data and the name of the invention up and down, and the output area 84 displays the data of the applicant of the patent document data. An output area indicated by each rectangle 83 (hereinafter referred to as “output area 83”) indicates a feature word information output area of the patent document data. It corresponds.

  In the present embodiment, feature word information composed of forward morphemes having a high feature word output order, that is, a high GF value, is displayed in order from the left side of the feature word output area.

  For example, the patent document data output to the output area 81a indicates the data of the patent ID “patent 1” of the document-specific output order information 70. Among the feature words 1 of “patent 1”, the feature word output order “Development” having the highest is displayed in the upper stage 85 of the output area 83 at the left end of the feature word output area, and the feature word 2 “perform” corresponding to the feature word 1 “development” is displayed in the lower stage 86. Feature word 1 and feature word 2 are displayed in the output area at the right end in order as the feature word output rank decreases.

  When feature word 1 and feature word 2 having the same feature word output order for each patent ID in document-specific output order information 70 are output, as shown in each rectangle 87 in FIG. Each combination of word 2 is displayed vertically.

<Discussion>
The feature word map 80 described above displays the feature words of each patent document data in a set unit of patent document data having high technical relevance, so that it is more relevant among patent document data groups belonging to a certain technical field. The technical characteristics of each high patent document data group can be shown to the user.

  Further, the feature word 1 and the feature word 2 of each patent document data are output according to the GF rank of the morpheme data constituting the feature word 1 in the feature word output area of each patent document data. To infer whether feature word 1 and feature word 2 ranked in the feature word output area are commonly used in the patent document data group to be individually analyzed or originally used Can do.

[Embodiment 3]
<Overview>
The information processing apparatus according to the present embodiment visualizes technical assets in a company to be analyzed. Specifically, the technical assets in the present embodiment are the technical elements that constitute the invention included in the patent document data group of the company, the product that is the subject of the invention constituted by each technical element, etc. In the embodiment, a first characteristic word (hereinafter, referred to as “technical element object word”) indicating a technical element constituting an invention included in a patent document data group, and a first characteristic word indicating an object of invention of each patent document data. Three feature words (hereinafter referred to as “product group target words”) are extracted, and a technical element keyword representing a technical factor latent in the invention of the patent document data group is expressed using the first feature word, and the patent document data group A product group keyword representing the product or the like is represented using a third feature word. In addition, relationship information indicating the relationship between the technical element keyword and the product group keyword, such as what technical factors are related to each product in the patent document data group, is output.
Details of the information processing apparatus in the present embodiment will be described below.

<Configuration>
A functional configuration of the information processing apparatus according to the present embodiment will be described.
FIG. 32 shows a functional configuration diagram of the information processing apparatus according to the present embodiment.
Hereinafter, each part of the information processing apparatus 100 will be described with reference to the same figure, but the configuration denoted by the same reference numerals as those of the first and second embodiments described above is the same as that of the first and second embodiments. Therefore, the description is omitted.

  The information processing apparatus 100 includes a storage unit 2, an input unit 3, a display unit 4, and a control unit 110. The control unit 110 includes an input reception unit 101, a data acquisition unit 102, a morpheme analysis unit 111, and features. A word extraction unit 112, a factor analysis unit 113, a factor specification unit 114, a cluster specification unit 115, a keyword generation unit 116, and an output control unit 117 are included.

  The control unit 110 is realized by a CPU and a memory such as a ROM and a RAM, and has a function of controlling each unit of the information processing apparatus 100 when the CPU reads and executes a program stored in the ROM.

Hereinafter, each part of the control part 110 is demonstrated.
The morpheme analysis unit 111 receives the patent document data group to be analyzed from the data acquisition unit 102, and whether or not the description format of each claim data of the claim data in each patent document data of the patent document data group is a predetermined format. In accordance with the application, the morpheme is detected from the specified part of each claim data, or the invention data described as the name of the invention of all the claim data and the patent document data, and the part of speech is associated with the detected morpheme It has a function of generating and storing part-of-speech information by number.

  Here, the predetermined portion includes a first predetermined portion (hereinafter referred to as “technical element target portion”) in each claim data in the claim data of each patent document data, and the claim range data. And a second predetermined portion (hereinafter referred to as “product group target portion”) in the first claim data described as claim 1.

  When each claim data in each patent document data is described in a predetermined format, the morpheme analyzer 111 reads the character string of the technical element target part (hereinafter referred to as “technical element target data”) and the product. Morphological analysis is performed on the character string of the group target portion (hereinafter referred to as “product group target data”), and the first morpheme and the second morpheme are detected by each morpheme analysis process. If each claim data of the patent document data is not in a predetermined format, a morpheme analysis is performed on each claim data of the patent document data and the name data of the invention to detect the first morpheme and the second morpheme.

  The predetermined format is, for example, a Jepson type description format such as “..., characterized by ...”. For each claim data, the morpheme analysis unit 111 refers to “in” (hereinafter referred to as “first character string”) and “characteristic” (hereinafter referred to as “second character string”). ) Is included, and the technical element target part is the “to do” part between the first character string and the second character string, and the product group target part is the first part of the first claim. The part of “***” written after the second character string.

  The feature word analysis unit 111 uses the first morpheme whose part of speech is the first case particle for each claim data of each piece of patent document data of the part number information by application number generated by the morpheme analysis unit 111, as in the first embodiment. For each of the first morphemes detected before the first morpheme (hereinafter referred to as “front first morpheme for each first case particle”), the first front part of the predetermined part-of-speech in which the detection order is continuous. It has a function of generating technical element target words by combining morphemes, and sending technical element target word information indicating each generated technical element target word to the factor analysis unit 113. In addition, the feature word analysis unit 111 sequentially generates clauses by combining the second morpheme based on the part of speech of the second morpheme for each claim data of each patent document data of the part number of part information by application number, and the patent The product group target word is generated by combining the clauses containing the second case particles with the phrase generation order continuing in order from the last phrase in the document data, starting with the last phrase generation order, and the generated product group target word and the product group target word The product group target word information indicating the application number of the patent document data corresponding to is sent to the cluster specifying unit 115.

  In this embodiment, the first case particle is “no” and “is”, the second case particle is “no”, and the predetermined part of speech is “noun” “unknown word”. . In addition, each clause generated for each patent document data is stored in association with the generation order in the patent document data.

Next, the factor analysis unit 113 will be described.
The factor analysis unit 113 reads the patent document data table, the part-of-speech information by application number, and the technical element target word information, and calculates the TF (Term Frequency) value of each technical element target word in all the claim data of each analyzed patent document data. Derived and has a function of generating document vector information of each technical element target word having each value as a component obtained by dividing each TF value by the total of all TF values of the patent document data. Further, the factor analysis unit 113 has a function of performing the following factor analysis using document vector information of each technical element target word using each technical element target word as an observation variable. The factor analysis in the present embodiment is performed using statistical analysis software such as SPSS (registered trademark) or R.

(I) With respect to a group of patent document data to be analyzed (number of patent document data I), n factors (1st factor to nth factor) with the technical element target words (n) of each patent document data as observation variables Is set as the initial factor.
(II) Based on the above settings, the factor loading for each factor of each technical element subject word is calculated using the SMC method and the principal factor method.
(III) Among the above factors, factors whose eigenvalues are equal to or greater than a predetermined threshold are extracted as target factors (N) of the analysis target patent document data group. In this embodiment, a factor having an eigenvalue of 1 or more is extracted.
(IV) For the target factor, the factor load matrix is obtained by rotating the factor axis using the varimax method.
(V) The factor score of each analysis target patent document data is calculated using the factor load matrix of each technical element target word calculated in (IV) above.
In addition, the factor analysis unit 113 further transmits the target factor information indicating the target factor to the factor specification unit 113 and the keyword generation unit 116, and the factor load amount and factor score calculated by the above (IV) and (V). It has a function of storing factor load amount calculation result information indicating each calculation result and factor score calculation result information.

Next, the function of the factor specifying unit 114 will be described.
The factor specifying unit 114 receives the information indicating the target factor sent from the factor analysis unit 113, and in the calculation result information of the factor load amount, the factor load amount of each technical element target word is the target factor having the first threshold value or more. A function to send the element information to be assigned to the keyword generation unit 116, which is specified as the element to be attributed to the element subject word, and to indicate the factor to be assigned to each technical element object word, and each analysis target patent document in the factor score calculation result information A function for specifying a target factor having a data factor score equal to or higher than a second threshold as an attribution target factor of the analysis target patent document data, and storing document attribution target factor information indicating an attribution target factor of each analysis target patent document data. Have. In this embodiment, it is assumed that the first threshold value is 0.2 and the second threshold value is 1.0 and stored in the ROM in advance.

  The cluster identification unit 115 receives product group target word information from the feature word extraction unit 112, and for each product group target word, in the product group target part of the first claim data of the analysis target patent document data group or the name data of the invention Function to obtain the DF value of the product group target word, TF value in each product group target word of each second morpheme of part-of-speech information by application number, and IDF (Inverse Document Frequency) of each second morpheme in all product group target words The document vector of the analyzed patent document data whose component is the value obtained by multiplying the TF value of each second morpheme and the IDF value is generated, and the document vector information by application number indicating each document vector is generated as a keyword A function of sending to the unit 116.

  Further, the cluster identification unit 115 is a document vector of product group target words having a DF value equal to or greater than a predetermined value among the product group target words of each analysis target patent document data (hereinafter referred to as “high DF document vector”). A clustering processing function for calculating a similarity between them and extracting a cluster, a document vector of a product group target word having a DF value smaller than the predetermined value (hereinafter referred to as “low DF document vector”), and the extraction The degree of similarity with each document vector belonging to each cluster is calculated, the function of assigning the low DF document vector to the cluster including the document vector having the highest similarity with the low DF document vector, and each analysis target patent document data It has a function of storing cluster information indicating a cluster to which it belongs and sending the cluster information to the keyword generating unit 116.

  Note that the similarity in the present embodiment is obtained by the cluster specifying unit 115 calculating cosine values between document vectors, and the cluster extraction is performed by sequentially clustering the document vectors having the maximum similarity as one group. Is generated by calculating the similarity between the document vectors not belonging to the clusters and the clusters or the clusters, and including the unaffiliated document vectors in each cluster using the longest distance method.

  The keyword generation unit 116 receives the target factor information indicating the target factor from the factor analysis unit 113 and the attribution target factor information indicating the attribution target factor of each technical element target word from the factor specifying unit 114, and the factor of each technical element target word Based on the load amount calculation result information, among the technical element target words belonging to each target factor, a technical element keyword is generated by combining technical element target words with a factor load of the third threshold or more, and the generated target It has a function of storing technical element keyword information for each factor. Further, the keyword generation unit 116 uses the function of receiving the cluster information and the document vector information by application number from the cluster specifying unit 115 and the document vector of the patent document data belonging to each cluster of the cluster information, and calculates the centroid vector of the cluster. A product group of analysis-target patent document data having a function of calculating and calculating a similarity between the centroid vector and each document vector belonging to the cluster, and a document vector corresponding to a predetermined rank or higher in descending order of similarity in the cluster A function for generating a product group keyword indicating the cluster by combining the target words, a function for storing the product group keyword information for each generated cluster, and a function for sending the technical element keyword information and the product group keyword information to the output control unit. Have. In the present embodiment, for example, it is assumed that the third threshold is stored in advance in the ROM as 0.2.

  The output control unit 117 receives the technical element keyword information and the product group keyword information from the keyword generation unit 116, and for each attribution target factor of the patent document data belonging to each cluster, based on the application number attribute information and the patent document data information. A function for counting the number of cases and generating the number-by-factor factor-specific information, a function for calculating the total evaluation value for each attribution target factor of patent document data belonging to each cluster, and generating the cluster-by-factor factor-specific evaluation value information, Based on the technical element keyword information and the product group keyword information, the number of cases by the cluster-specific factor number information and the first relation information in which the technical element keyword and the product group keyword corresponding to the number are associated are displayed on the display unit 4. Function, each evaluation value of evaluation value information by factor for each cluster, technical element keyword and product group keyword corresponding to the evaluation value A function of causing the display unit 4 to display the second relation information associated with the mode.

Here, examples of the first relation information and the second relation information will be described with reference to FIG.
FIG. 46A shows an example of the first relation information in the present embodiment. In the first relation information 630 of FIG. 46, product group keywords 1 to M (632) are the products of the product group keyword information. Group element keywords 1 to N (631) indicate element element keywords of the element element keyword information, and each product group keyword and each cell corresponding to each element element keyword are patent document data. The number of cases is shown. For example, the cell 633 indicates that the number of patent document data belonging to the product group keyword 2 and having the technical element keyword N as the attribution target factor is five.

  FIG. 46B shows an example of the second relation information in the present embodiment. The second relation information 640 in FIG. 46 includes the technical element keywords 1 to N (631) on the X axis and the Y axis. Is a three-dimensional graph in which product group keywords 1 to M (642) and an evaluation value 643 are set on the Z axis. For example, a column 644 in the figure shows the total value of the evaluation values of patent document data belonging to the product group keyword 1 and having the technical element keyword 1 as an attribution target factor.

<Data>
Hereinafter, the data structure stored in the storage unit 2 or the memory of the information processing apparatus 100 according to the present embodiment will be described.

FIG. 33A shows the configuration and data example of the patent document data table.
The patent document data table 510 is read when the data acquisition unit 102 acquires the applicant's patent document data received by the input reception unit 101 as an analysis target of the present embodiment.

  The patent document data table 510 in the figure stores an application number 511, an applicant 512, an invention name 513, a claim 514, and an evaluation value 515 in association with each other.

  The application number 511 is the application number of the patent application relating to each patent document data, the applicant is the name of the applicant of the patent application, and the name of the invention 513 is the name of the invention in the application specification of the patent application. The claims 514 are data described as claims or claims in the patent application, and all claims data of the patent application are stored for each claim. ing. The evaluation value 515 is data indicating the evaluation of the invention according to the patent application preset by the user by a predetermined calculation method.

FIG. 33B shows the configuration and data example of the part number part-of-speech information table by application number.
The part number part-of-speech information table 520 is generated when the morphological analysis unit 111 performs morphological analysis on the data of the claim 514 of the patent document data table 510 or the data of the invention name 513 of each patent document data to be analyzed. Is done.

  The part number part-of-speech information table 520 of FIG. 5 stores an application number 521, a first ID 522, a first morpheme 523, a part of speech 524, a second ID 525, a second morpheme 526, and a part of speech 527 in association with each other.

  The application number 521 is the application number of the patent document data subjected to morphological analysis, and the first ID 522 is the claim of the morpheme detected in the technical element target portion in each claim data of the claim 514 of the patent document data. This is data indicating the claim number of the data and the detection order in the claim data. For example, when the first ID 522 is “1-1”, it indicates that the detection order is the first in the first claim.

  The first morpheme 523 is morpheme data detected from the technical element target portion of each claim data of the patent document data, and the part of speech 524 is a part of speech corresponding to each morpheme of the first morpheme 523. The second ID 525 is data indicating the detection order of the morphemes detected in the product group target portion in the first claim data of the claim 514 of the patent document data, and the second morpheme 526 is the patent document data. Morpheme data detected from the product group target portion of the first claim data, and the part of speech 527 is a part of speech corresponding to each morpheme of the second morpheme 526.

FIG. 34A shows the configuration and data example of the technical element target word-specific document vector information.
The technical element target word-specific document vector information 530 shown in FIG. 5 includes the technical element target word information generated by the feature word extraction unit 112 when the factor analysis unit 113 performs factor analysis of the patent document data group to be analyzed. It is generated based on all the claim data of the patent document data group.

  The technical element target word-specific document vector information 530 stores an application number 531 and each technical element target word 532 in association with each other.

  The application number 531 is the application number of the patent document data to be subjected to factor analysis, and the technical element target word 532 is a claim of all patent document data for each technical element target word generated by the feature word extraction unit 112. This is a component of the document vector of the technical element target word obtained by dividing each TF value of the technical element target word in the data by the total TF value for each patent document data.

FIG. 34B shows the configuration and data example of document vector information by application number.
The document number-specific document vector information 540 in FIG. 11 is obtained by using the product group target word generated by the feature word extraction unit 112 and the first of each patent document data when the cluster specifying unit 115 clusters the patent document data group to be analyzed. It is generated based on the claim data or the name data of the invention.

Application number-specific document vector information 540 stores application number 541, product group target word 542, DF 543, storage box and the like 544 in association with each other.
The application number 541 is the application number of each patent document data to be analyzed, the product group target word 542 is a product group target word extracted by the feature word extraction unit 112 in the patent document data, and the DF 543 is a patent DF value data of each product group target word in the product group target portion of the first claim data of the document data group, and the storage box etc. 544 is added to each TF value in each product group target word of each second morpheme. A value obtained by multiplying the IDF value of the second morpheme in the product group target word is shown.

  The DF 543 is used as a reference value for the cluster identification unit 115 to distinguish between a high DF document vector and a low DF document vector.

FIG. 35B shows the configuration and data example of factor load amount calculation result information.
The factor load amount calculation result information 550 shown in the figure is generated when the factor analysis unit 113 calculates the factor load amount of each technical element target word using each document vector of the technical element target word-specific document vector information 530. .

The factor load amount calculation result information 550 stores the technical element target word 551 and the first factor to the Nth factor 552 in association with each other.
The technical element target word 551 is a technical element target word extracted from the analysis target patent document data group, and the first factor to the Nth factor 552 are target factors, and correspond to each technical element target word and each target factor. Each cell stores a factor load value for the target factor of the technical element target word.

FIG. 35C shows the configuration and data example of factor score calculation result information.
The factor score calculation result information 560 in the figure is generated when the factor score of each patent document data is calculated based on the factor load amount calculation result information 550.

The factor score calculation result information 560 is stored in association with the application number 561 and the first to Nth factors 562.
The application number 561 is the application number of each patent document data subject to factor analysis, the first factor to the N-th factor 562 are target factors, and each cell corresponding to each application number and each target factor has its application number. The factor score value for the target factor is stored.

FIG. 36 (a) shows the configuration and data example of attribution information by application number.
The application number-specific attribution information 570 in FIG. 8 stores cluster information of clusters to which each patent document data belongs when the cluster identification unit 115 performs clustering on the patent document data group to be analyzed, and the factor identification unit 114 stores the cluster information. Document attribution target factor information is stored when the attribution target factor of each patent document data is specified.

The application number-specific attribution information 570 stores an application number 571, a cluster number 572, and an attribution target factor 573 in association with each other.
Application number 571 is the application number of each patent document data to be analyzed, cluster No. 572 is the cluster number of the cluster to which the patent document data belongs, and attribution target factor 573 is attributed to the patent document data. The target factor information is shown.

FIG. 36B shows the configuration and data example of the technical element keyword information.
The technical element keyword information 580 in FIG. 5 is generated by the keyword generation unit 116 based on the target factor information received from the factor analysis unit 113, the attribution target factor information received from the factor specifying unit 113, and the factor load amount calculation result information 550. It is stored when a technical element keyword indicating each target factor is generated.

The technical element keyword information 580 stores the target factor 581 and the technical element keyword 582 in association with each other.
The target factor 581 indicates each target factor of the target factor information received by the keyword generation unit 116 from the factor specifying unit 113, and the technical element keyword 582 combines technical element target words having the target factor as an attribute target factor. Indicates the technical element keyword. For example, the technical element keyword 1 is obtained by inserting a comma between technical element target words “alloying elements”, “alloying elements”, “flakes”, and “particles” and combining them. Other technical element keywords are also generated in the same manner, but for the sake of convenience of description, expressions such as technical element keyword 2, technical element keyword 3,.

FIG. 36C shows the configuration and data example of product group keyword information.
The product group keyword information 590 shown in the figure is stored when the keyword generation unit 116 generates a product group keyword indicating each cluster based on the cluster information of the document vector information 540 by application number and the attribution information 570 by application number. The

The product group keyword information 590 stores a cluster number 591 and a product group keyword 592 in association with each other.
Cluster No. 591 indicates the cluster number of each cluster in the cluster information, and the product group keyword 592 is a product group generated by combining product group target words in patent document data belonging to the cluster. Indicates a keyword. For example, product group keyword 1 is generated by combining the product group target words of “slide fastener” and “slider fastener slider” in the same manner as the above technical element keyword, and the other product group keywords are the same. is there.

FIG. 37A shows the configuration and data example of the cluster-specific factor number information.
The number-by-cluster factor number information 610 in FIG. 11 is based on the application number attribution information 570 and the patent document data table 510. The output control unit 117 uses the attribution information of the patent document data belonging to each cluster as the first relation information. It is generated when the number of patent document data is output every time.

The cluster-specific factor number information 610 stores clusters 1 to M612 and first to Nth factors 611 in association with each other.
Cluster 1 to cluster M 612 are each cluster of cluster information of attribution information 570 by application number, and first factor to N factor 611 indicate each target factor, for example, indicated by cluster 1 and N factor. The cell 613 stores the number of patent document data belonging to the cluster 1 and belonging to the Nth factor.

FIG. 37B shows a configuration and data example of cluster-based factor-by-factor evaluation value information.
The cluster-based factor-specific evaluation value information 620 in the figure is based on the application number attribution information 570 and the patent document data table 510, and the output control unit 117 uses the second relation information as the attribution object of the patent document data belonging to each cluster. Generated when outputting the total evaluation value of patent document data for each factor.

The cluster-specific factor-by-factor evaluation value information 620 stores clusters 1 to M622 and first to N-th factors 621 in association with each other.
Cluster 1 to cluster M 622 are each cluster of the cluster information of attribution information 570 by application number, and first factor to N factor 621 indicate each target factor, for example, indicated by cluster 2 and N factor. The cell 623 stores the total evaluation value of patent document data belonging to the cluster 2 and belonging to the Nth factor.

<Operation>
Hereinafter, the operation of the information processing apparatus 100 according to the present embodiment will be described.
FIG. 38 shows an operation flow showing the overall operation of the information processing apparatus 100. Hereinafter, description will be given with reference to FIG.

  In step S <b> 1100, the input receiving unit 101 of the information processing apparatus 100 receives a designation input by the applicant of the patent document data group to be analyzed from the user via the input unit 3, and the input receiving unit 101 receives the data acquiring unit 102. Sends analysis target information indicating the specified applicant.

  The data acquisition unit 102 reads the patent document data table 510 from the storage unit 2, reads patent document data corresponding to the analysis target information received from the input reception unit 101, and reads the analysis target patent document data to the morpheme analysis unit 111. The group information is transmitted (step S1200).

  The morpheme analysis unit 111 performs morpheme analysis processing using the information of the patent document data group received from the data acquisition unit 102 (step S1300).

Here, the details of the morphological analysis processing will be described with reference to FIG.
The morpheme analysis unit 111 extracts each claim data in the claim data 514 of the patent document data for each patent document data of the patent document data group to be analyzed (step S1310).

  For each claim data extracted in step S1310, the morpheme analyzer 111 determines whether or not the description format of the claim data matches a predetermined format (step S1320). Whether or not it conforms to a predetermined format is determined by whether or not a predetermined character string is included. For example, in the claim data shown in FIG. 35A, in the underline 50A, the first character string “characterized by” and the second character string “characterized by the underline 50C” are included in the claim data. Therefore, it is determined that the claim data conforms to a predetermined format.

  In step S1320, when the morpheme analysis unit 111 determines that the description format of the claim data matches the predetermined format (step S1320: Y), the morpheme analysis unit 111 determines the technical element target portion of the claim data. Are extracted (step S1330). In the case of the claim data in the example of FIG. 35A described above, each character string sandwiched between the first character string of the underline 50A and the second character string of the underline 50C, that is, the character string portion indicated by the underline 50B is Each character string of the underline 50B, which is a technical element target portion, is extracted.

  Subsequently, the morphological analysis unit 111 determines whether or not the claim data is the first claim data (step S1340), and determines that the claim data is the first claim data (step S1340). : Y), the morpheme included in the data of the product group target part in the claim data is detected, and each detected morpheme is extracted as the second morpheme (step S1350). In the case of the claim data shown in FIG. 35A, the character string after the second character string of the underline 50C, that is, the part of the character string indicated by the underline 50D is the product group target part, and each of the underline 50D A second morpheme is extracted from the character string.

  The morpheme analyzer 111 detects the morpheme included in the technical element target data of the claim data extracted in step S1330, and extracts the detected morpheme as the first morpheme (step S1360).

  Subsequently, the morpheme analysis unit 111 associates the first morpheme and the second morpheme corresponding to the first morpheme and the second morpheme of the claim data extracted in steps S1350 and S1360, and detects the first morpheme and the first morpheme in the order detected in the claim data. The first ID 522 and the second ID 525 indicating the detection order are attached to each of the two morphemes, the part-of-speech information 520 by application number is stored in the memory, and end information indicating that the morpheme analysis processing is ended is sent to the feature word extraction unit 112. (Step S1370).

  In step S1320, if the morpheme analysis unit 111 determines that the description format of the claim data is not a predetermined format (step S1320: N), the morpheme analysis unit 111 uses all character strings of the claim data as a technology. A morpheme is detected as element target portion data, and the detected morpheme is extracted as a first morpheme (step S1380). Subsequently, the morpheme analysis unit 111 detects a morpheme from the name 513 of the invention corresponding to the application number of the claim data in the patent document data table 510, and extracts the detected morpheme as a second morpheme (step S1390). The above-described processing in step S1370 is performed on the extracted first morpheme and second morpheme.

Returning to FIG. 38, each processing after step S1400 will be described.
When the feature word extraction unit 112 receives the end information from the morpheme analysis unit 112 in step S1300, the feature word extraction unit 112 uses the morpheme data stored in the first morpheme 523 and the second morpheme 526 of the part number part-of-speech information 520 in the memory. Thus, the technical element target word in the analysis target patent data group and the product group target word for each analysis target patent data are generated (step S1400).

Details of the processing in step S1400 will be described with reference to FIG.
The feature word extraction unit 112 reads the part-of-speech information by application number 520 from the memory (step S1410), and stores the part-of-speech information in the part-of-speech 524 for each claim number of each application number stored in the application number 521 of the part-of-speech information by application number 520. For the first morpheme 523 in which the first case particle of “no” or “is” is stored, the front first morpheme of the first morpheme is extracted (step S1420).

  The feature word extraction unit 112 generates the technical element target word by combining the first morpheme of the predetermined part of speech with the continuous first ID 522 among the first morpheme for each claim data of each application number extracted in step S1420. (Step S1430).

  Subsequently, the feature word extraction unit 112 sequentially generates a phrase by combining the second morphemes for each application number of the part-of-speech information 520 by application number, and associates the generation order with each generated phrase (step S1440). .

  The feature word extraction unit 112, for the clauses generated for each application number in step S1440, the phrase generation order is continuous from the last phrase generation order, and the second case particle in the part of speech 527 of the second morpheme included in the phrase " A product group target word is generated by combining up to the phrase including "". Further, the technical element target word information of the technical element target word generated in step S1430 is sent to the factor analysis unit 113, and the product group target word information indicating the product group target word is sent to the cluster specifying unit 115 (step S1450).

  Returning to FIG. 38, upon receiving the product group target word information from the feature word extraction unit 112, the cluster specifying unit 115 performs clustering of the analysis target patent document data group using each product group target word information of the product group target word information. This is performed (step S1500).

Details of the clustering process will be described below with reference to FIG.
In step S1510 of FIG. 41, the cluster specifying unit 115 reads the patent document data table 510 in the storage unit 2 and the part-of-speech information 520 by application number in the memory.

  The cluster specifying unit 115 sets the description format of the first claim data included in the claims 514 of the patent document data table 510 of the analysis target patent document data group for each product group target word of the product group target word information as a predetermined format. If the description format of the first claim data is not a predetermined format, the DF value of the product group target word in the invention name 513 is derived, and the DF value and the DF value The application number of the patent document data corresponding to and the product group target word are associated with each other and stored in the document vector information 540 by application number (step S1520).

  The cluster specifying unit 115 calculates the TF value in the product target word corresponding to the application number of each second morpheme for each application number of the part-of-speech information 520 by application number, and the second morpheme in all product group target words. An IDF value is calculated (step S1530).

  The cluster specifying unit 115 multiplies the TF value of each second morpheme calculated for each application number calculated in step S1530 and the IDF value of the second morpheme as a component of the document vector of the product group target word of the application number. It is stored in the document vector information 540 by application number (step S1540).

  Subsequently, the cluster specifying unit 115 refers to the DF 543 of the document vector information 540 by application number stored in step S1530, extracts a high DF document vector, and obtains a cosine value between the extracted high DF document vectors. Similarity between product group target words is calculated, and clusters are extracted using the longest distance method (step S1550).

  The cluster specifying unit 115 extracts the low DF document vector by referring to the DF 543 of the document vector information 540 by application number, and calculates the similarity between the document vector belonging to each cluster extracted in step S1550 and each low DF document vector. Then, by assigning the low DF document vector to a cluster including the document vector having the highest similarity with the low DF document vector, the belonging cluster of all product group target words is determined. The cluster specifying unit 115 stores the cluster information in which the application number corresponding to each product group target word and the cluster number of the belonging cluster are associated with each other in the application number belonging information 570, and sends the cluster information to the keyword generating unit 116 ( Step S1560).

  38, when the factor analysis unit 113 receives the technical element target word information from the feature word extraction unit 112 in step S1400 in step S1600, the analysis target patent document data of each technical element target word in the technical element target word information. The factor analysis of the patent document data group to be analyzed is performed using the appearance frequency in.

Details of the operation in step S1600 will be described below with reference to FIG.
The factor analysis unit 113, for each technical element target word of the technical element target word information received from the feature word extraction unit 112, in the claim 514 of the patent document data table 510 corresponding to the application number of each analysis target patent document data A TF value is derived (step S1610), and a value obtained by dividing the TF value of the technical element object word for each application number derived in step S1610 by the total TF value of the application number is used as a document vector component of each technical element object word. The document is stored in the technical element target word-specific document vector information 530 (step S1620).

  Subsequently, the factor analysis unit 113 performs each factor analysis using each document vector of the document vector information 530 for each technical element target word, with each technical element target word as an observation variable and the number of technical element target words as an initial factor number. Then, the factor loading of each technical element target word is calculated, and a factor having an eigenvalue of 1 or more is extracted as the target factor. Further, the factor analysis unit 113 calculates a factor load matrix by rotating the factor axis for the target factor, and calculates a factor score of each analysis target patent document data using the factor load matrix (step S1630).

  The factor analysis unit 113 sends the target factor information extracted in step S1630 to the factor specifying unit 114, stores the factor load amount after rotation obtained in step S1630 as factor load amount calculation result information 550, and each analysis target patent. The calculation result of the factor score of the document data is stored as factor score calculation result information 560 (step S1640).

  Returning to FIG. 38, in step S1700, the factor specifying unit 114 determines each technique based on the target factor information, factor load amount calculation result information 550, and factor score calculation result information 560 received from the factor analysis unit 113 in step S1600. The target factor to which each of the element target word and each analysis target patent document data belongs is specified.

Details of step S1700 will be described below with reference to FIG.
In step S <b> 1710 of FIG. 43, upon receiving the target factor information from the factor analysis unit 113, the factor specifying unit 114 reads the factor load amount calculation result information 550 and the factor score calculation result information 560.

  For each technical element target word of the technical element target word 551 in the factor load amount calculation result information 550, the factor specifying unit 114 is a target factor whose factor load amount of the target factor corresponding to the technical element target word is equal to or greater than the first threshold. Is specified as the attribution target factor of the technical element target word, and the technical element attribution target factor information in which the technical factor target word to which the target factor belongs is associated with the target factor is sent to the keyword generation unit 116 ( Step S1720).

  Subsequently, for the patent document data of each application number of the application number 561 of the factor score calculation result information 560, the factor specifying unit 114 applies the target factor whose factor score of the target factor corresponding to the application number is the second threshold value or more. The document attribution target factor information in which the application number with the target factor as an attribution destination is identified and associated with the target factor is sent to the keyword generation unit 116 (step S1730). .

  Returning to FIG. 38, in step S1800, the keyword generation unit 116 uses the technical element target word to indicate each target factor based on the technical element attribution target factor information and the document attribution target factor information received from the factor specifying unit 114. An element keyword is generated, and a product group keyword indicating each cluster is generated using the product group target word.

Details of step S1800 will be described with reference to FIG.
Upon receiving the cluster information sent from the cluster specifying unit 115 in step S1500 and the technical element attribution target factor information and document attribution target factor information sent from the factor specifying unit 114 in step S1700, the keyword generating unit 116 receives the factor load The amount calculation result information 550 is read (step S1810).

  The keyword generation unit 116 combines the technical element target words whose factor loading is equal to or larger than the third threshold in the factor loading calculation result information 550 among the technical element target words belonging to each target factor of the technical element attribution target factor information. Then, a technical element keyword indicating the target factor is generated for each target factor. Further, the keyword generating unit 116 sends the technical element keyword information 580 to the output control unit 117 and stores the technical element keyword information 580 (step S1820).

  The keyword generating unit 116 obtains the center-of-gravity vector of the cluster using the document vector of the application number-specific document vector information 540 of the application number of the patent document data belonging to each cluster of the cluster information received in step S1810, and the cluster The degree of similarity between the cluster and the patent document data belonging to the cluster is calculated by calculating the cosine value of the document vector and the center-of-gravity vector of each application number belonging to (Step S1830).

  The keyword generating unit 116 combines the product group target words corresponding to the patent document data having document vectors of a predetermined rank or higher in descending order of similarity between each cluster calculated in step S1830 and the patent document data belonging to the cluster. A product group keyword indicating the cluster is generated. Further, the keyword generation unit 116 sends the product group keyword information 590 to the output control unit 117, and stores the product group keyword information 590 (step S1840).

  Returning to FIG. 38, in step S1900, the output control unit 117 generates and outputs the relationship information between each product group keyword and each technical element keyword generated by the keyword generation unit 116 in step S1800.

Details of step S1900 will be described below with reference to FIG.
45, the output control unit 117 receives the product group keyword information 590 and the technical element keyword information 580 sent from the keyword generation unit 116 in step S1800. In step S1920, the output control unit 117 The above-mentioned attribution information by application number 570 and the patent document data to be analyzed are read out.

  The output control unit 117 counts the number of patent document data belonging to each cluster in the attribution number-specific attribution information 570 for each factor to be attributed, and the counted number of each factor for each target factor as cluster-specific factor number information 610. Store (step S1930).

  Subsequently, the output control unit 117 reads the evaluation value of the analysis target patent document data read in step S1910, and calculates the total evaluation value for each attribution target factor of the patent document data belonging to each cluster in the application number attribution information 570. The calculated evaluation value sum for each target factor of each cluster is stored as cluster-specific factor evaluation value information 620 (step S1940).

  The output control unit 117 reads the technical element keyword indicating the number of cases in the cluster-specific factor number information 610 and the target factor corresponding to the number of cases from the technical element keyword information 580, and selects the product group keyword indicating the cluster corresponding to the number of cases. Read from the product group keyword information 590, and display the first relation information (FIG. 46A) in which the number of cases, the technical element keyword corresponding to each number of cases, and the product group keyword are associated with each other (step S1950). .

  Subsequently, the output control unit 117 reads out from the technical element keyword information 580 the technical element keyword indicating each evaluation value of the cluster-specific evaluation value information 620 and the target factor corresponding to the evaluation value, and corresponds to the evaluation value. The product group keyword indicating the cluster is read from the product group keyword information 590, and the second relation information (FIG. 46B) in which each evaluation value is associated with the technical element keyword corresponding to each evaluation value and the product group keyword is displayed. It is displayed on the part 4 (step S1960).

<Calculation processing of evaluation values by factor by cluster>
Next, the cluster-based factor-by-factor evaluation value calculation process described above will be described. The cluster-based factor-based evaluation value calculated here is referred to as a “cluster score”.
FIG. 47 is a flowchart illustrating a procedure of cluster score calculation processing according to the embodiment of this invention. The cluster score calculation process is executed by the output control unit 117 of the information processing apparatus 100 or a cluster score calculation unit (not shown).
It is assumed that the patent score (PS) for each patent document belonging to each cluster and factor is calculated before performing the processing of FIG. Then, in the memory (or storage unit 2) of the information processing apparatus 100, for each information (gazette number) identifying the patent document, the “patent score (PS)” of the patent document and the patent are abandoned. It is assumed that information (hereinafter referred to as “PS information”) in association with “abandonment information (including information indicating whether rejection has been confirmed)” indicating whether or not is stored is stored. . The procedure for calculating the patent score (PS) will be described with reference to FIGS.

Specifically, the information processing apparatus 100 receives a request for cluster score calculation processing from the user via the input unit 3 (S2010). Note that when the user requests the cluster score calculation process, the user also designates a category to be calculated.
As a classification to be calculated, for example, a classification for each attribution target factor of patent document data belonging to each cluster in the application number-specific attribution information 570 is designated.

Next, the information processing apparatus 100 acquires a patent score (PS) of a patent document belonging to a category (cluster and factor) that is a cluster score calculation target received in S2010 (S2020).
Specifically, the information processing apparatus 100 uses “information in which patent documents are associated with each cluster and each factor (application number-specific attribution information 570)” and “PS information” stored in the memory. The “patent score (PS)” and “abandonment information” of the patent documents belonging to the clusters and factors to be calculated are acquired.

  Next, the information processing apparatus 100 uses the “patent score (PS)” and “abandonment information” of the patent documents belonging to the acquired cluster and factors to be calculated, and the patent score (PS) that has not been abandoned. Each of the standard values is obtained (S2030).

Specifically, the information processing apparatus 100 refers to the “waiver information” and, among the patent documents belonging to the designated cluster and factor, patent documents that have not been surrendered (including applications pending at the Patent Office) Specify a patent score (PS).
The information processing apparatus 100 obtains a standard value for the specified patent score (PS) in a population (for example, a patent document that has not been surrendered in the analysis target document group subjected to cluster extraction processing). More specifically, the information processing apparatus 100 obtains a standard value for each identified patent score (PS) using the following (Equation 1) and the identified patent score (PS).

In (Equation 1) shown below, it is assumed that there are “m” patent scores (PS) of patent documents that have not been waived in the population, and a subscript i is added to the patent score (PS), and “PSi ( 1 ≦ i ≦ m (m is an integer of 1 or more)) ”.
Further, in (Expression 1), the standard value of “patent score PSj” of each patent document j belonging to a specific cluster and factor among the PSis of m patent documents is obtained.

  Next, the information processing apparatus 100 obtains the total value of the standard values of the patent score PSj greater than or equal to the threshold value among the standard values of the patent scores PSj of the patent documents belonging to the specific cluster and factor obtained in S2030, and the total The value is set as the “cluster score” of the cluster and the factor (S2040). In this step, the information processing apparatus 100 obtains the maximum value among the standard values of the patent scores PSj of the patent documents belonging to the specific cluster and factor obtained in S2030.

Specifically, the information processing apparatus 100 uses the following (Equation 2) and the standard value of the patent score (PSj) obtained in S2030, and the “cluster score” for the cluster and factor specified by the user. Is calculated. Further, the information processing apparatus 100 selects the maximum (MAX) standard value from the standard values of each patent score PSj obtained in S2030, and sets the selected standard value as the maximum value in the cluster and factor.
In (Expression 2), among the standard values of the patent scores PSj obtained in S2030, the number of standard values of the patent score PSj equal to or greater than the threshold is “n” in the cluster and factor. In (Expression 2), as an example of the threshold value PSstd, the average of the standard values of each patent score PSi obtained in S2030 (0 according to [Expression 1]) is used.

When the cluster score is calculated by the information processing apparatus 100, the process proceeds to S1960 (output) in FIG.
In the flow of FIG. 47, the cluster score for one cluster and factor is calculated, but this is merely an example. When a request for calculating the cluster scores of a plurality of clusters and factors is received, the processing of S2020 to S2040 is performed for each cluster and factor, and the cluster score and the maximum value are obtained for each cluster and each factor.

  In S1960 of FIG. 45, the output device 4 outputs the cluster score obtained in S2040. Alternatively, the output device 4 outputs the maximum value of the cluster and factor together with the cluster score.

As described above, in the present embodiment, the cluster score is calculated using the patent score (PSi) of a patent document that has not been waived. The reason for this is as follows. For example, when a company tries to evaluate patents for each technical field, the number of patent documents classified into a certain technical field (cluster and factor) is very large, but many of them are abandoned ( Or an application for which a decision of rejection has been finalized). In such a case, if an application that has already been abandoned (or an application for which refusal has been finalized) is included in the evaluation of a patent in that technical field, the technical field that does not hold many patent rights will be highly evaluated. Therefore, proper analysis is not possible.
Therefore, in the present embodiment, the cluster score is calculated using the patent score (PSi) of a patent document that has not been abandoned so as to improve the accuracy of the score.

  Further, in the present embodiment, when calculating the standard value of the patent score (PSi), a general standard value is multiplied by a coefficient instead of a simple standard value ((Equation 1) is multiplied by 10). ing). This is for facilitating the discrimination between the obtained standard values. In (Equation 1), it is 10 times, but it is only an example.

In this embodiment, only the standard value of the patent score PSi exceeding the threshold is used for calculating the cluster score. This is to alleviate the influence of the number of patent documents that the cluster score value receives.
For example, it is assumed that a cluster score is obtained for each cluster and each factor, and the obtained cluster score is compared to analyze the technical tendency for each cluster and each factor. In this case, if the threshold value is not taken into consideration as in the present embodiment, the cluster score having a large number of applications and the cluster score values of factors tend to be too high, and there is a possibility that analysis with high accuracy cannot be performed.
Certainly, if the patents in a particular technical field are extracted without excess or deficiency and used as the analysis target document group (population), the number of applications for each cluster and factor itself should be considered as a sufficiently significant value. Can do. However, if an analysis target document group (population) is extracted by an arbitrary method that is not so, if the number of applications for each cluster and factor is limited, there is a possibility that a highly accurate analysis cannot be performed. There is.
In addition, when the focus is on selecting important elements from a group of documents to be analyzed (population) including a huge number of patents, the “individual importance” is more than the “large number of patents with low individual importance”. In some cases, it is preferable to focus on those that include “high patents”.
For this reason, in the present embodiment, only the standard value of the patent score PSi that is equal to or greater than a predetermined value is used, and a high cluster score is given only to clusters and factors that include important patents that are equal to or greater than the predetermined value. In this way, the accuracy of the cluster score was increased.
In particular, for example, when a patent score is standardized so that the average becomes 0 and a standard value equal to or higher than the average (0) is aggregated to obtain a cluster score, not only the patent score value below the average can be discarded, but also the average Even if there are many patent scores in the vicinity, the influence on the value of the cluster score is small, and if there is something that is far from the average and is high, the value of the cluster score is greatly affected. Therefore, it is possible to further reduce the influence of the number of cases included in the technical elements and accurately extract the technical elements including the patents with high importance.

  In this embodiment, when calculating the cluster score, since the patent scores are excluded excluding publications with less than average, the clusters and factors that occupy many publications with less than average, or all publications are less than average. For clusters and factors, the cluster score is close to 0 or 0. Therefore, the contrast between the clusters and the factors becomes clear, and as a result, the order and evaluation between the clusters and the factors can be easily grasped visually.

In the present embodiment, the average of the population is used as the threshold, but the present invention is not particularly limited to this. For example, an average of the standard values of the patent score PSi in the patent group of the specific applicant or other threshold values determined by other users may be set in the information processing apparatus 100.
In the present embodiment, the standard value of the patent score PSi is used, but the present invention is not limited to this. For example, the influence of the number of cases can be mitigated even when only non-standardized patent scores PSi are added that are greater than or equal to a predetermined value.

Further, according to the present embodiment, when the cluster score is presented to the user, the highest standard value of the patent score (PSj) of the patent document classified into the cluster and the factor can be presented. As a result, the user can grasp which technical element (cluster and factor) the highly evaluated patent is included in. Accordingly, even if the evaluation value as a whole of the technical elements (clusters and factors) is low, the user can grasp the technical elements (clusters and factors) including the highly evaluated patent.
For example, assume that a company obtains a cluster score for each cluster and each factor of the company (applicant) in an attempt to evaluate a patent for each technical field. In this case, by presenting the maximum value for each cluster and factor, it is possible to grasp which technical field of the company has a strong patent.

<About patent score (PS)>
Next, the patent score (PS) used for calculating the cluster score in the embodiment will be described with reference to FIGS. 48 to 51.
The patent score (PS) calculation process is performed by the output control unit 117 of the information processing apparatus 100 or a patent score calculation unit (not shown), but is not particularly limited thereto.
Another computer having a CPU (Central Processing Unit), a memory, and the like may perform the patent score calculation process. In this case, a program for realizing the patent score calculation function (PS calculation program) is stored in another computer. Then, the CPU of another computer executes the “PS calculation program”, thereby calculating the patent score PS and generating the above-described PS information. The information processing apparatus 100 acquires PS information generated by another computer and stores it in the memory.

(Data structure)
First, the data structure used for calculating the patent score PS will be described.
The storage unit 2 stores patent data (electronic data indicating a patent gazette) and patent attribute information. The electronic data indicating the patent publication includes at least the patent data ID (gazette number, etc.), the application date, and the bibliographic information such as the IPC code.
The patent attribute information includes progress information 300 of the patent document (information such as whether priority is claimed or the number of citations in examination of other patent applications), and content information 400 (number of claims, Information such as the number of specifications). Hereinafter, the data structure of the progress information 300 and the content information 400 will be described.

First, an example of the data structure of the progress information 300 is shown in FIG.
FIG. 48 is a diagram schematically showing an example of the data configuration of the progress information used in the present embodiment.
As shown in the figure, the progress information 300 includes a field 301 for registering “patent data ID (gazette number, etc.)”, a field 302 for registering “the number of days elapsed from the filing date”, and “examination request date”. A field 303 for registering "elapsed days since", a field 304 for registering "elapsed days since registration date", a field 305 for registering information indicating whether "divisional application" exists, A field 306 for registering information indicating the presence / absence of “early examination”, a field 307 for registering information indicating the presence / absence of “trial decision of appeal”, and information indicating the presence / absence of “opposition maintenance decision” are registered. A field 308 for registering, a field 309 for registering information indicating the presence / absence of “invalidation trial decision”, and information indicating the presence / absence of “priority claim” A field 310 for registering information indicating the presence / absence of “PCT application”, a field 312 for registering information indicating the presence / absence of “packaging browsing”, and “number of times cited” And a field 313 for registering the information to be shown constitutes one record. The progress information 300 includes a plurality of records.

  Here, “elapsed days from application”, “elapsed days from examination request”, and “elapsed days from registration date” are information relating to the period of the corresponding patent data. “Elapsed days from application” is the application date, “Elapsed days from examination request” is the application examination request date, and “Elapsed days from registration date” is the evaluation date (calculation of patent score). The number of elapsed days up to a predetermined date close to the evaluation date is calculated and stored in the storage unit 2. “Elapsed days from examination request” for a patent application that has not yet been requested for examination of application is NULL, and “elapsed days from registration date” for a patent application that has not yet been set and registered is NULL.

  Among the progress information 300, “divisional application”, “accelerated examination”, “approval appeal decision”, “opposition maintenance decision”, “invalidity decision maintenance decision”, “packaging browsing”, “priority” This is information indicating the presence or absence of a predetermined action on the data. "Divisional application" is whether the divisional application has been filed based on the patent application, "Rapid examination" is whether the patent application has been expedited, and Whether an appeal against a decision to reject the application has been requested and whether a patent trial decision has been made in that trial, whether or not the opposition maintenance decision has been made, whether or not a patent opposition has been made and a maintenance decision has been made on the patent, The “invalidation trial maintenance decision” is whether the patent invalidation trial has been requested for the patent, and whether the appeal has been rejected in the trial, “priority” is the priority claim based on the previous patent application etc. Whether or not the patent application is an international application based on the Patent Cooperation Treaty and whether or not it is a domestic application. Based on whether it is, respectively given 1 for example if the predetermined action has been performed, if not been given a 0, for example.

Next, the data structure of the content information 400 is shown in FIG.
FIG. 49 is a diagram schematically illustrating an example of a data configuration of content information used in the present embodiment.

As shown in the figure, the content information 400 includes a field 401 for registering “patent data ID (gazette number, etc.)”, a field 402 for registering “number of claims” of the patent data, and “claim One record is composed of a field 403 for registering the “average number of characters” and a field 404 for registering the “number of specifications” of the patent data. The content information 400 includes a plurality of records.
Here, the “number of claims” is information indicating the number of claims of the patent application, and the “average number of characters of the claim” is the average number of characters (or the number of words) per claim of the patent application. It is information to show. The “number of specification pages” is information indicating the number of specification pages or publication pages of the patent application. Such information is extracted from published patent gazettes and other patent data of each patent application.

(Patent score calculation process)
Next, description will be made with reference to FIG. FIG. 50 is a flowchart showing a procedure of a patent score calculation process according to this embodiment.

As shown in FIG. 50, the information processing apparatus 100 receives an IPC code input from the user and acquires patent data (electronic data indicating a patent publication) (S400).
Specifically, when receiving an IPC code input from a user, the information processing apparatus 100 accesses the storage unit 2 and acquires patent data classified into the IPC code. The patent data includes bibliographic information such as the filing date information and priority date information of the patent application (only when priority is claimed).

Next, the information processing apparatus 100 uses the application date information or the priority date information among the bibliographic information of the acquired patent data, and converts the patent data every predetermined period (in this embodiment, every application year, the priority date is (S500).
Next, the information processing apparatus 100 calculates an evaluation value of each patent data (S600). Details of this processing will be described with reference to FIG.

FIG. 51 is a flowchart showing details of processing for calculating an evaluation value of each patent data according to the present embodiment.
The information processing apparatus 100 acquires the progress information 300 and the content information 400 for the patent data belonging to the group generated by the classification of S210 (S610). Specifically, the information processing apparatus 100 uses the patent ID (gazette number or the like) included in the bibliographic information of the acquired patent data to store the progress information 300 and the content information 400 stored in the storage unit 2. From the above, the progress information 300 and the content information 400 associated with the patent ID of the acquired patent data are acquired.
Here, in FIG. 51, it is assumed that the obtained one group is composed of J patent data, and a subscript j (j = 1, 2,..., J) is used to distinguish each of the J cases. .
When J patent data is acquired, using the J patent data progress information 300 and content information 400, “total value for J of the evaluation item corresponding data” used in S6302 to S6304, which will be described later, etc. Is obtained in advance.

  Next, the variable j is set to 1 (S620), and the evaluation raw score of the patent data j is calculated as follows.

  First, information registered in each field of the progress information 300 is used as an evaluation item, and I evaluation items i (i = 1, 2,..., I) are evaluated in advance for each evaluation item. A calculation method is selected (S6301).

  There are the following three evaluation point calculation methods in the present embodiment. That is, for information registered in the fields 305, 306, 307, 308, 309, 310, 311 and 312, S6302 [Presence / absence type] is selected as information indicating the presence / absence of a predetermined action on the patent data. For fields 302, 303, and 304, S6303 [time decay type] is selected as information related to the period of the patent data. In the field 313, S6304 [number-of-times] is selected as information indicating the number of times the patent data is cited.

  When the evaluation score calculation method is selected, the evaluation score of the patent data j is calculated for each of the I evaluation items i (S6302, S6303, S6304).

(Evaluation score for presence / absence type)
For the evaluation item i for which S6302 [presence / absence type] is selected, the evaluation score is calculated by the following [Equation 3].

  Here, the “existence / non-existence data of the evaluation item i” arranged in the numerator is, for example, “1” if the divisional application has been filed as described above, and “0” if it has not been made.

In the denominator, the positive square root of the in-group total value of the above “evaluation item i presence / absence data” is arranged. Therefore, the denominator is large when there are many patent data corresponding to the evaluation items in the group, and the denominator is small when there are only a few patent data corresponding to the evaluation items in the group. Patents with fewer evaluation items (such as “Invalidation Trial Maintenance Decision”) than patents with a higher number of evaluation items (such as “Bag Viewing”) will be maintained after patent registration (In general, a high maintenance rate is considered to indicate a high economic value commensurate with the maintenance cost (patent fee)), and thus each evaluation item is automatically weighted. In addition, since it is tabulated in groups for each predetermined period, for example, older patents have more progress information added, and new patents that have just been published often do not yet have progress information added. It can alleviate the tendency for new patents to be given low ratings for that reason.
The attribute information of the patent data is useful for relative evaluation within the analysis target population, but proper evaluation cannot be performed if the patent applications or patent rights in the analysis target population are treated equally. According to the present embodiment, the analysis object population including patent applications or patent rights at different periods is classified by classifying the analysis object population into groups for each period and using the value obtained for each classified group as a denominator. Appropriate relative assessment is possible within the population.
For example, in a certain technical field, the former value is often higher between one value in a simultaneous group with few patent applications and one value in a simultaneous group with many patent applications. On the other hand, for example, a patent application that has passed several years is more likely to be given progress information, such as a request for browsing, than a patent application that has just been published. It is an error to underestimate a patent application that has just been made. For example, if only a few of the patent applications in the same period group have been requested to be browsed, the patent application that has received the request for browsing is a patent application with a particularly high degree of attention and should be highly evaluated. On the other hand, if there are a large number of requests for inspection among patent applications in the same period group, the patent application that received the request for inspection should be highly evaluated just because it was requested for inspection. is not.
According to the present embodiment, the value obtained using the patent attribute information of each patent data belonging to each group and the value obtained using the patent attribute information of each patent data belonging to the group are determined for each group. The evaluation score is calculated by multiplying the sum of the values by the value of the decreasing function. According to this structure, the value which considered the relative positioning of each patent data in each group can be calculated | required as an evaluation value. As a result, the lower the total value of the numerical information based on the progress information in the simultaneous group, the higher the weight, and conversely the lower the higher the total value, the lower the weight, so that Appropriate evaluation is possible.

(Calculation of evaluation points for time decay type)
For the evaluation item i for which S6303 [Time decay type] is selected, the evaluation score is calculated by the following [Equation 4].

  “Exp (− (Min (elapsed time, year limit)) / year limit)” placed in the numerator here is the “number of days elapsed since the request for examination”. ”Or“ year ”, which is the value obtained by dividing the smaller value by“ year ”and multiplying by −1 and the power of the Napier number e. The “year” is the maximum number of years from the filing date until the expiration of the patent right (20 years under the current Japanese law). The same formula is used for “elapsed days from registration date”, and “year” is the maximum number of years from the filing date to the expiration of the patent term (20 years under the current Japanese law). The same formula is used for the “elapsed days from the filing date”, but the “year” is the number of years from the filing date to the application examination request deadline (3 years in the current Japanese law). According to this, while the elapsed time is short, the value of the numerator is a value close to Exp (0) = 1. However, when the elapsed time ≧ years, the value of Exp (−1) = 1 / e. To fall. The advantage of using an exponential function is that a depreciation effect on the value can be introduced and that the evaluation value distribution can be eliminated and a smooth distribution can be achieved. “Elapsed days from request for examination”, “Elapsed days from application date”, and “Elapsed days from registration date” are basic evaluation items applicable to many patents. Tying can be avoided.

  The denominator has the same formula as the above S6302 [Presence / absence type], but the “days since examination request” is, for example, 1 if an application examination request is made for the patent application, and if not, for example 0 Are summed within the group to obtain a positive square root. For the “elapsed days from the date of registration”, the denominator is a value obtained by adding a value of 1 within the group by taking the positive square root by adding 1 if the patent application has been registered for patent right setting and not being registered. . Since all patent data falls under “Elapsed days since filing”, the value of the denominator is equal to the positive square root of the number of patent data in the group, assuming that the evaluation data of the relevant evaluation item is 1. . In any case, the denominator is large when there are many patent data corresponding to the evaluation items in the group, and the denominator is small when there are only a few patent data corresponding to the evaluation items in the group. As described above, “Elapsed days from request for examination”, “Elapsed days from application date”, and “Elapsed days from registration date” are basic evaluation items applicable to many patents. Tends to be small.

The evaluation score calculated in S6303 [time decay type] is further corrected by content information.
In the following, the content information 400 shown in FIG. 49 is used.
When the evaluation is based only on the progress information, there is a possibility that the patent application or the patent right shortly after the publication of the application or the registration of the patent right does not have the progress information expected to be granted in the future and the evaluation cannot be performed correctly. Therefore, in order to correct this, content information is added to the evaluation based on the progress information. However, the content information tends not to have a high correlation with the maintenance rate as the progress information, and if the content information is inadvertently added, the accuracy of the evaluation may decrease.
Therefore, in order to keep the influence of the content information small in the evaluation of a patent with sufficient progress information and to effectively reflect the content information in the evaluation of a patent with insufficient progress information, this S223C [time decay type] Only the evaluation score calculated in (5) is multiplied by the correction coefficient based on the content information.
As described above, according to the present embodiment, regardless of whether the application is old or new, it is possible to add the content information of each patent data to the information related to the period having characteristics that are easily given to any patent data. As a result, it is possible to perform appropriate evaluation even for patent data consisting of a new application to which little progress information is given.

Specifically, for each evaluation point in [Equation 4],
a ₁ × a ₂ × a ₃
here,
a ₁ = 2 ^1/3 (when the average number of characters per claim is below average) or 2 ^-1/3 (when the average number of characters per claim is above average)
a ₂ = 2 ^1/3 (when the total number of pages is above average) or 2 ^-1/3 (when the total number of pages is below average)
a ₃ = 2 ^1/3 (when the number of claims is within an average value ± 1 standard deviation) or 2 ^−1/3 (when the number of claims is outside the above range)
Multiply By setting the maximum values of a ₁ , a ₂ , and a ₃ to 2 ^1/3 , the correction is limited to a ₁ × a ₂ × a ₃ as the maximum value. In the above embodiment, the value of a ₁ × a ₂ × a ₃ is set to 2 at the maximum.

(Evaluation score calculation for the frequency type)
For the evaluation item i for which S6304 [number-of-times] is selected, an evaluation score is calculated by the following [Equation 5].

Here, “f (quotation) × log (n _j +1)” arranged in the numerator is the weight of the logarithm of the value obtained by adding 1 to the “cited count n _j ” for the “cited count”. Quoting). According to the verification by the present inventors, it has been found that the maintenance rate of the patent right changes depending on the number of citations as well as the presence or absence of citations. Since the increase gradually shows a tendency to peak, the logarithm is taken.

In the denominator, the positive square root of the total value in the group of “f (quotation) × log (n _j +1)” is arranged. Accordingly, the denominator is large when there are many patent data cited in other applications in the group, and the denominator is small when there are only a few patent data cited in other applications in the group.

具体的な重みとしては、他社引用の場合のｆ（引用_other）と、自社引用の場合のｆ（引用_self）との比を、１：２とした。In the numerator and denominator of [Formula 5], an arbitrary positive number can be used as the weight f (quotation), but the number of times cited in other patent applications (number of times other companies cited) n _jother and other patents of the company It is distinguished from the number of times cited in the application (in-house citation number) n _jself, and a different weight is _assigned to each logarithm. In this case, instead of the above [Equation 5], the following [Equation 6] is used.

As a specific weight, the ratio of f (quoting _other ) in the case of _other company citations and f (quoting _self ) in the case of company citations was set to 1: 2.

The number of times cited is highly correlated with the value of a patent. Furthermore, according to the verification by the present inventors, the number of times cited in the examination of patent applications of other companies (citation of other companies) and the number of times cited (in-house quotation) in examinations of other patent applications of the company are the latter. Was found to be significantly higher in correlation with patent value. The invention cited in the examination of other patent applications of the company is presumed to be due to the fact that it is often the basic invention that is the core in the implementation technology of the company. And while recognizing that the company has already applied for such a basic invention, there is a high possibility that the company has applied for the improved technology and built a strong patent portfolio.
According to this embodiment, it is possible to appropriately evaluate a patent application or a patent right by considering the number of citations separately from other company citations and company citations, and reflecting the latter number more largely in the evaluation value.

この式に示されるように、評価素点は、Ｉ個の評価点の二乗和の正の平方根、又は０となる。評価素点が０となるのは、審査請求期限までに出願審査請求しなかった場合、出願を取下げ又は放棄した場合、拒絶査定が確定した場合、その他特許出願が失効した場合と、異議申立による取消決定や無効審判による無効審決が確定した場合、特許権を放棄した場合、特許権の存続期間が満了した場合、その他の特許権が消滅した場合である。これらの情報も各特許データの経過情報から読み取り、該当する場合は評価素点を０とする。
上述のようにＳ６３０３〔時間減衰型〕で算出された評価点に対しては、内容情報による補正を行う。具体的には、「審査請求からの経過日数」、「出願日からの経過日数」、「登録日からの経過日数」に基づき上述の［数４］で算出された評価点にそれぞれ上述のａ_１×ａ_２×ａ_３を乗算した上で、［数７］に従い二乗和の平方根をとる。(Calculation of evaluation raw score)
When the evaluation score of the patent data j is calculated for all the evaluation items i (i = 1, 2,..., I), the evaluation score of the patent data j is calculated based on the evaluation score of the following [Expression 7]. ] (S640).

As shown in this equation, the evaluation raw score is a positive square root of the sum of squares of I evaluation points, or 0. The evaluation score is 0 because the application request is not requested by the deadline for requesting examination, the application is withdrawn or abandoned, the decision of refusal is finalized, other patent applications have expired, The case where the decision of revocation or the trial for invalidation by the trial for invalidation is finalized, the patent right is abandoned, the duration of the patent right expires, or the other patent right is extinguished. These pieces of information are also read from the progress information of each patent data, and the evaluation raw score is set to 0 when applicable.
As described above, the evaluation score calculated in S6303 [time decay type] is corrected by the content information. Specifically, each of the evaluation points calculated in the above [Equation 4] based on “the number of days elapsed since the request for examination”, “the number of days elapsed since the application date”, and “the number of days elapsed since the registration date” _After multiplying by ₁ × a ₂ × a ₃ , the square root of the sum of squares is taken according to [Equation 7].

As a method of calculating an evaluation raw score from an evaluation point i based on a plurality of evaluation items, there is a method of calculating a sum of each evaluation point i (simple sum method). However, according to this calculation method, since the evaluation of a patent to which a lot of historical information having a correlation with the patent maintenance rate (economic value) is given is calculated high, the sum of the evaluation points i should be used as an evaluation raw score. Is reasonable at first glance, but the evaluation score of a patent that has been granted a lot of historical information that does not have a high correlation with the maintenance rate (a lot of low evaluation points are added) has a very high correlation with the maintenance rate Care should be taken because it may exceed the evaluation score of a patent to which a small amount of progress information is granted.
As one method for solving this problem, there is a method in which the maximum value among the evaluation points i is used as an evaluation raw score (maximum value method). However, according to this calculation method, especially when investigating the correlation between certain historical information and the maintenance rate of a group of patents, when investigating the correlation regardless of what other historical information is given, maintaining a certain patent Since the rate is expected to be best expressed by the maintenance rate of the historical information having the highest maintenance rate, it is reasonable to use the maximum value of the evaluation point i as an evaluation raw score. If the maximum value is the same in the two patents, no superiority or inferiority is given. Furthermore, when the maximum value method is used, it is not possible to carry out an evaluation that takes into account the viewpoints of three different entities of the applicant, the JPO, and competitors, and only the viewpoints of any one of those entities The viewpoint of the remaining subject cannot be reflected in the evaluation of patent data.
The above-described method for taking the square root of the sum of squares can be said to be a method that combines the advantages of the simple sum method and the maximum value method. That is, by taking the square root of the sum of squares, when there is a high evaluation point i in I evaluation items i related to a certain patent data j, the high evaluation point i greatly affects the evaluation raw score. The evaluation points other than the evaluation item having a high evaluation point i are also evaluation raw points that are somewhat considered. Therefore, a high evaluation score is given to patent data j that corresponds to multiple items such as “early examination”, “opposition to maintain opposition”, and “invalidation maintenance decision” that tend to be high. be able to.
As described above, in this embodiment, patent evaluation is performed in consideration of all evaluation points calculated according to the type of patent attribute information (S630, S640). As a result, it is possible to evaluate the value of patent data from multiple aspects.

(Calculation of evaluation value)
When the evaluation raw score is calculated, its logarithm is calculated and used as the evaluation value of the patent data j (S650).
The evaluation value calculated based on the progress information or content information is given a high value for a few patent applications or patent rights whose unique progress or content can be read, but for many other patent applications or patent rights. Is often given a low value. Accordingly, looking at the distribution of the number of evaluation values, the number of patent applications or patent rights with high evaluation values is a few and sparse distribution, and the number of patent applications or patent rights with low evaluation values is a dense distribution.
In such a case, the average value (arithmetic average value) is greatly influenced by a small number of patent applications or patent rights with high evaluation values, so care must be taken when evaluating by comparison with such average values. It becomes. In addition, for example, when comparing two patent applications or patent rights that have obtained high evaluation values, even if it appears that there is a large difference in evaluation values, it may not be a significant difference in practice. is there.

Next, it is determined whether or not evaluation values have been calculated for all patent data j (S660). If not calculated (S660: NO), the process proceeds to S670, the variable j is set to j + 1, and the process returns to S630. The evaluation value is calculated for the following patent data.
When the evaluation values are calculated for all the patent data j (S660: YES), the evaluation value calculation processing for the patent data belonging to the group ends.
As described above, in the present embodiment, a plurality of patent data having different characteristics are evaluated in consideration of the characteristics for each technical field and each filing time. As a result, the value of patent data can be more appropriately evaluated.

The evaluation value calculation processing from S610 to S670 is executed for all the groups t obtained by classifying the patent data acquired in S400 in S500.
When the evaluation values are calculated for all the groups t, the process returns to FIG. 50, and the deviation value in the analysis target population acquired in S400 is calculated as the patent score PS based on the evaluation values (S700). This deviation value also enables relative comparison of patent data between different technical fields that are difficult to compare (comparison with a population to be analyzed separately selected by different IPCs in S400). is there.

In this embodiment, since the cluster score is calculated based on the patent score PS obtained by the above procedure, there are the following advantages compared to the above embodiment.
Specifically, in the above embodiment, the patent score PS that is the basis of the cluster score takes into account the weight according to the type of progress information. Since the cluster score is obtained using the patent score PS, a score with higher accuracy is calculated in this embodiment.
According to the patent score of the present embodiment, the analysis target population is classified into groups for each period, and the values obtained for each classified group are used as denominators, thereby including patent applications or patent rights at different periods. Appropriate relative evaluation is possible within the analysis population.
Therefore, it is possible to reduce the possibility that a high evaluation value is calculated for the cluster and the cluster score of factors in which many patent data whose applications are old are classified.

<Discussion>
As described above, the information processing apparatus according to the present embodiment can output the first relation information or the second relation information in which the technical element keyword and the product group keyword are associated with each other. It is possible to grasp the relationship between R & D technology and products using that technology. Specifically, since it is possible to confirm whether or not technical elements common to mutually independent product groups are used, it is possible to prevent duplicate research and development. In addition, for example, it is possible to check the usage status of each technical element to the product, such as the state where the technical elements embodied in many products and the technical elements that are not commercialized are unevenly distributed. It is possible to improve the efficiency of research and development by effectively utilizing the technical assets of the company.

<Supplement>
The information processing apparatus according to the present invention has been described using the first embodiment and the second embodiment, but the information processing apparatus according to the present invention is not limited to this, and includes the following modifications.

   (1) In the cluster generation processing of the first embodiment described above, if the number of patent document data of the generated cluster is less than or equal to the predetermined number, the generated cluster is canceled and each patent belonging to the cluster is released. It is also possible to calculate the degree of similarity with other clusters for the document data, and make the patent document data belong to the cluster with the highest degree of similarity.

   (2) In Embodiment 1 described above, the longest distance method is used for the cluster generation processing. However, the present invention is not limited to this, and the cluster generation processing is performed by a method such as the shortest distance method or the Ward method. You may go.

  (3) In the embodiment described above, in the morpheme combination process of the front morpheme and the back morpheme for each case particle, the morpheme until the morpheme other than the first classification or the second classification appears in the part of speech is combined in the detection order. However, in the case of the front morpheme for each case particle, the front morpheme whose part of speech belongs to the first classification is extracted once from among the front morphemes, and the extracted front morpheme has a detection rank from the front morpheme immediately before the case particle. The forward morphemes may be combined as long as they are continuous. Similarly, for the back morphemes of each case particle, the back morpheme whose part of speech belongs to the second classification is once extracted from the back morphemes, and the detection order of the extracted back morphemes continues from the back morpheme immediately after the case particles. As long as backward morphemes are combined, they may be combined.

  (4) In the second embodiment described above, the map generation unit 108 has been described as outputting all the feature words 1 and 2 in the document-specific feature word extraction list 40 (FIG. 24) to the feature word map. For example, an upper limit number of feature word information to be output to the feature word output area may be set in advance. In that case, the feature word information to be output to the feature word output area is restricted by allowing the user to select a predetermined upper number of the feature word output rankings 75 of the document-specific output order information 70 (FIG. 25C).

  In the second embodiment, the description has been made assuming that each output area of the feature word output area and the feature word output order correspond to each other one by one. It is good also as making a continuous output order correspond.

  (5) Further, each output area of the feature word output area of the feature word map of the second embodiment described above may be output with a color according to the output order.

  (6) In the first embodiment described above, the summary data included in each patent application data is used as the problem information. However, in the specification data included in the patent application data, the “problem to be solved by the invention” is described. It is good also as using the data described as "."

  In the first and second embodiments, the data described as “Claim 1” in the scope of claims is used as the solution information. However, the data of other claims may be used. However, all the data in the claims may be used.

  (7) In the embodiment described above, when the morpheme combination process is performed on the forward morpheme for each case particle, the forward morpheme corresponding to any of the noun, unknown word, symbol, and adjective whose part of speech is the first classification is detected in the order of detection. Although described as a combination, for example, a front morpheme whose part of speech is only a noun may be combined, or a noun and an unknown word, or a noun and an unknown word or a symbol or an adjective front morpheme may be combined.

  Further, in the second embodiment, when the backward morpheme combination processing is performed on the backward morpheme for each case particle, the backward morpheme corresponding to any of the noun, unknown word, symbol, adjective and verb whose part of speech is the second classification is detected. Although described as being combined sequentially, the noun and verb, or the noun and verb or unknown word or symbol or adjective backward morpheme may be combined in the same manner as the above combination of the front morpheme.

  (8) In the second embodiment described above, when the user designates a cluster in the problem / solution matrix output in the first embodiment, feature word information in each patent document data belonging to the cluster is generated, and the feature word Although described as generating and outputting a map, it accepts input specifying one or more patent document data from the user, generates feature word information for the specified patent document data, and specifies the specified patent document. It is good also as producing | generating and outputting the feature word map about data.

  (9) In the above-described embodiment, it has been described that when extracting a feature word, attention is given to predetermined case particles “O” and “GA”. It is also possible to focus on the case particles.

  (10) In the above-described embodiment, when performing morphological analysis processing, such as “above” and “above” are frequently used in the document but are not characteristic words in the document (hereinafter referred to as “unnecessary words”). .) Has also been described as generating part-of-speech information by performing morphological analysis processing, but an unnecessary word list is stored in the information processing apparatus in advance, and words registered in the unnecessary word list are not included in the part-of-speech information. May be.

  In addition, when generating a feature word, when punctuation included in a symbol whose part of speech is the first classification is a combination target of morpheme combination processing, morphemes excluding punctuation may be combined.

  (11) The output position of the task word and the solution word in the first embodiment described above is the position where the task words having the highest similarity and the solution words are adjacent to each other on each axis of the matrix by the linked list generation process described above. Although it has been described that the other task words and solution words are arranged in descending order of similarity in the two directions of each axis starting from the combination of task words and solution words having the highest similarity, respectively. For example, apply the above linked list generation process only for the axis of the task word and place each task word, and for the axis of the solution word, in the order of the appearance frequency of the solution word, such as descending order or ascending order of the GF value of the solution word Each solution word may be arranged.

  According to this, the user can check at a glance the solution words and feature words that are frequently used or not used in the designated patent document data group without referring to the values of the cells in the matrix. The technical tendency of the entire designated patent document data group can be estimated. In addition, since the task words on the task word axis are arranged by the linked list generation process described above, the user can easily confirm the appearance frequency of the solution words and feature words between similar task words on the matrix. Therefore, it is easy to grasp the difference and similarity of technical means for solving each problem between similar patent document data groups.

  (12) In the first embodiment described above, when there are a plurality of combinations of the reference linked lists generated by the linked list generation process, a plurality of linked lists are generated by linking the reference linked lists of each set, When connecting a plurality of linked lists, it has been described that each linked list is linked in descending order of similarity between each reference node in each linked list, but ascending order of similarity between each reference node in each linked list The linked lists may be linked according to the similarity between the head nodes or the tail nodes of the linked lists.

  (13) In the above-described second embodiment, when the output positions of the feature word 1 and the feature word 2 are determined, the GF value of the front morpheme constituting the feature word 1 is used. The DF value in the specified patent document data group of the morpheme may be used, and the feature word 1 is higher as the detection position in the document data to be analyzed of the front morpheme constituting the feature word 1 or the feature word 1 is higher A weight corresponding to the detected position of the feature word 1 or the front morpheme constituting the feature word 1, such as weighting, may be used.

  (14) In the above-described embodiment, the description has been made on the assumption that patent application data filed in Japanese is used as the analysis target document. For example, a technology such as a technical paper in which the subject matter or problem of the document is clearly indicated. Document data or document data described in a markup language such as HTML (HyperText Markup Language) may be used, or patent application data described in Korean whose grammar is similar to Japanese may be used.

  (15) In the above-described embodiment, the data acquisition unit 102 has been described as acquiring patent document data to be analyzed from the patent document data group stored in advance in the storage unit 2 of the information processing apparatus 1. However, for example, patent document data may be acquired from an external terminal such as a server connected to the information processing apparatus 1 via a network. In the above-described embodiment, the information processing apparatus 1 has been described as receiving information indicating a patent document data group to be analyzed from the user via the input unit 3 of the information processing apparatus 1. Information indicating patent document data to be analyzed may be received from a user via an external terminal such as a computer connected to the processing apparatus 1 via a network.

  (16) Further, the present invention may be the method described in the above embodiment, a computer program that implements these methods by a computer, or a digital signal composed of the computer program. Also good.

  Further, the present invention may be the computer program or the digital signal recorded on a computer-readable recording medium such as a hard disk, CD-ROM, or DVD, or the computer recorded on the recording medium. It may be a program or a digital signal.

In the present invention, the computer program or the digital signal may be transmitted via the Internet or an electric communication line such as a wireless or wired communication line.
(17) In the third embodiment described above, when determining whether the description format of each claim data of the claim data is a predetermined format, in the first character string “˜”, “and the second character string”. Although it has been described that it is determined whether or not the data of “characteristic” is included, for example, the first character string is “˜,” “,” and includes a reading mark such as “”. It may be a character string indicating a precondition, or when one claim data includes a plurality of first character strings, the same character string as the last character string in the claim data is If it is described immediately before the first character string described last in the claim data, the last first character string may be used as the determination criterion of the predetermined format.
(18) In the third embodiment described above, the factor analysis by the factor analysis unit 113 has been described as using statistical analysis software such as SPSS (registered trademark) or R. However, the initial setting of the factor analysis (I) is described above. If it is a program which performs factor analysis based on this, it will not be restricted to this. Further, the factor analysis unit 113 assumes a factor load matrix and a factor score matrix based on the setting conditions of the factor analysis (I), obtains a correlation matrix of variables based on the technical element target word-specific document vector information, Estimate commonality using the SMC method or MAX method, calculate the factor loading using the principal factor method or least squares method, determine the target factor based on the calculated factor loading, and It is also possible to calculate the factor load amount obtained by rotating the factor axis orthogonally or obliquely, and calculating the factor score using the factor load amount after the rotation and the correlation matrix.
(19) In the third embodiment described above, for each technical element keyword related to the product group keyword, the first relation information indicating the number of patent document data to which the product group keyword belongs as a cluster (FIG. 46 (a)). However, for each product group keyword, information indicating whether or not the product group keyword and each technical element keyword are related may be output. In this case, for example, the related technical element keyword is set to 1, and the unrelated technical element keyword is set to 0. For example, the related information is expressed using numerical values and symbols.
(20) In Embodiment 3 described above, the first relation information and the second relation information are output. However, the first relation information or the second relation information may be output according to the user's designation. Good.
(21) In Embodiment 3 described above, the first relationship information is represented in two dimensions, and the second relationship information is represented in three dimensions. However, any relationship information is represented in two dimensions and three dimensions. It is good as well.
(22) In addition, the patent document data table in the third embodiment described above is obtained by extracting data of some items included in each patent application data filed at the Japan Patent Office. It may be data.
(23) In the third embodiment described above, when the keyword generation unit generates the product group keyword, a predetermined rank or higher in descending order of the similarity between the centroid vector of the cluster and the document vector of the patent document data belonging to the cluster. The product group target words corresponding to the patent document data of the above are described as being combined. However, for example, the product group target words of the patent document data whose similarity is equal to or greater than a predetermined value are to be combined, and the similarity to the cluster Depending on the product group target words to be combined may be determined.
(24) In the third embodiment described above, the factor analysis unit calculates the TF value of each technical element target word in all the claim data of each analysis target patent document data as the total of all TF values of the analysis target patent document data. The description has been made assuming that the document vector component of each technical element target word is obtained by division. As described above, the method of dividing each TF value by the total of all TF values of each patent document data to be analyzed considers that the weight of the technical element target word is different depending on the number of characters of the claim data, that is, the request. This is an effective method when considering the fact that the weight of patent document data with a large number of characters in the term data is different from the weight of patent document data with a small number of patent documents data. You may use TF value of an element object word as a component of a document vector.
In the third embodiment, the TF value is used as the component of the document vector. However, each TF value of each technical element target word is multiplied by the IDF value of the technical element target word in all analysis target patent document data. The component of the document vector of the technical element target word may be obtained by using the appearance rate of the technical element target word such as the calculated value.

  The information processing apparatus according to the present invention is used to analyze document data such as technical papers and manuals in general industries such as industry and commerce, and to search for a document desired by a user, in order to achieve a certain purpose. can do.

Claims (27)

An information processing apparatus that performs morphological analysis processing on document data, detects morphemes in the document data, decomposes the document data into morpheme data, and analyzes the document data,
Storage means for storing the document data;
A feature word generating means for performing the morpheme analysis processing on the document data and generating a first feature word composed of the morpheme data based on a predetermined first rule;
An information processing apparatus comprising: an output unit configured to output information indicating a tendency of the document data using the first feature word generated by the feature word generation unit. The information processing apparatus further includes:
Part-of-speech information generation means for generating first part-of-speech information that associates each decomposed morpheme data, a predetermined part-of-speech corresponding to each piece of morpheme data, and detection rank information indicating the detection order of each piece of morpheme data;
In the case where the predetermined participle is included in the first part of speech information, the feature word generating unit includes, for each predetermined case particle, from the predetermined case particle out of the morpheme data of the first part of speech information. The information processing apparatus according to claim 1, wherein the first feature word is generated using forward morpheme data which is morpheme data detected before. The first rule is that, among the front morpheme data for each predetermined case particle in the first part-of-speech information, the part-of-speech is from the front morpheme data detected immediately before the predetermined case particle in the first part-of-speech information. The information processing apparatus according to claim 2, wherein the front morpheme data until the front morpheme data belonging to parts of speech other than one category are detected is combined in the order of detection. The information processing apparatus according to claim 3, wherein the first classification is a noun. The part-of-speech information generating means generates the part-of-speech information by using, as an unknown word, part-of-speech of morpheme data that does not correspond to any of the predetermined part-of-speech among the decomposed morpheme data,
The information processing apparatus according to claim 4, wherein the first classification further includes the unknown word. The information processing apparatus according to claim 5, wherein the first classification further includes a symbol and an adjective. The document data is patent document data including claim scope data described as claims,
The information processing apparatus according to claim 3, wherein the morphological analysis process targets the range data of the claims. The output means determines an output rank of the first feature word based on the appearance frequency of the morpheme data constituting the first feature word, and outputs the first feature word according to the output rank The information processing apparatus according to claim 1, wherein an output process of information indicating the tendency is performed. The information processing apparatus further includes:
It is determined whether or not the description format of the claim range data matches a predetermined format, and when it is determined that the description format matches the predetermined format, the claim format data is predetermined by the predetermined format. The information processing apparatus according to claim 8, further comprising: a specifying unit that specifies data in a range between the predetermined data as the morpheme analysis processing target. The patent document data is a plurality of patent document data, and each patent document data further includes problem information including a plurality of character string data as a problem of the document,
The information processing apparatus further includes:
Cluster classification means for classifying the plurality of patent document data into clusters based on problem information of each patent document data,
The feature word generation means further includes, among the front morpheme data of the predetermined case particle, the front morpheme data whose part of speech belongs to the first classification is a solution word candidate target in the claim data group of the plurality of document data A solution word that is extracted as data, calculates the appearance frequency of each of the solution word candidate target data in the plurality of patent document data, and extracts the solution word candidate target data having a predetermined rank or higher in descending order of the calculated appearance frequency as the solution word Extraction means;
A solution word appearance frequency calculating means for calculating an appearance frequency of each solution word in each cluster;
The first feature word includes the solution word, and among the combinations including the solution word, the appearance frequency in the plurality of patent document data satisfies a predetermined criterion,
The appearance frequency of the morpheme data constituting the first feature word is an appearance frequency calculated by the solution word extraction unit of the solution word included in the first feature word,
The output means further outputs the solution word corresponding to the first feature word and the appearance frequency of the solution word calculated by the solution word appearance frequency calculation means for each cluster. Item 9. The information processing apparatus according to Item 8. The information processing apparatus further performs a morphological analysis process on each piece of task information,
The part-of-speech information generation means further includes each morpheme data decomposed by morpheme analysis processing on the task information, a predetermined part-of-speech corresponding to each morpheme data, and detection order information indicating a detection order of each morpheme data. Generate second part-of-speech information
The cluster classification means calculates an appearance frequency in each patent document data for morpheme data in which the part of speech belongs to the first classification among the morpheme data of the second part of speech information, and each document data with each calculated appearance frequency as an element A plurality of patent document data is classified into clusters by calculating a similarity between the document vectors,
Furthermore, the detection order in each patent document data of each morpheme data in which the part of speech belongs to the first classification is specified based on the second part of speech information, and the combined morpheme data by combining the morpheme data having the specified detection order continues A task word extracting means for calculating the appearance frequency in each cluster of each generated combined morpheme data, and extracting each combined morpheme data in a descending order of the appearance frequency as a task word indicating the cluster. Prepared,
The information processing apparatus according to claim 10, wherein the output unit further outputs the task words indicating the clusters. The output means represents each task word and each solution word and the set of the first characteristic words corresponding to the solution word in each row and each column of the matrix, and as each element of the matrix The information processing apparatus according to claim 11, wherein a value of an appearance frequency of the solution word corresponding to the task word and the solution word indicated by the row and the column configuring the element is represented. The output means further includes:
A cluster document vector indicating the document vector of the cluster is generated using the document vector of each document data generated by the cluster classification means, and a distance between the clusters is calculated using the cluster document vector of each cluster. By doing so, an inter-cluster similarity calculation means for obtaining the similarity between the clusters,
A vector of each solution word is generated using the value of the appearance frequency for each solution word calculated by the solution word frequency calculation means as an element, and a distance between each solution word is calculated using the vector of each solution word And a means for calculating the similarity between the solutions to obtain the similarity between the solutions,
For each similarity calculated by the inter-cluster similarity calculation means and the inter-solution similarity calculation means, a combination having the maximum similarity is extracted and included in each of the extracted cluster and solution word pairs. Each set of reference concatenations of clusters and solution words with each element as a reference element is generated, and one of the two connection directions starting from the reference element of the reference connection is defined as the reference of the cluster and the solution word. Set one reference connection of each set of connections, set the other connection direction to the other reference connection of the set, and set the cluster set and solution set in descending order of the respective similarities of the cluster and solution Are sequentially selected, and according to a predetermined condition, each of the selected combination elements is connected to each of the reference connections, and according to the result of each connection process, all the clusters and the output order of the solution words are performed. And an output order determining means for determining,
The predetermined condition includes a first condition and a second condition. In the first condition, in each combination of the selected cluster and the solution word, one element included in the combination is connected to any reference connection. The second condition is that any of the elements included in the combination is unconnected,
In the connection process, when any one element of the selected combination satisfies the first condition, the combination process is not performed in the connection direction set in the reference connection including the connected elements satisfying the first condition. A first connection process for connecting the elements of the connection, and when each element of the selected combination satisfies the second condition, a first reference connection set is generated using each element of the combination as a new reference element 2 connection process and a third connection process for connecting the reference connections of each set according to the number of elements connected to each of the reference connections and connecting the connected reference connections of each set based on a predetermined rule Including
The output is in accordance with the output order of each cluster and each solution word determined by the output order determining means, each task word, each solution word, and each solution word corresponding to the first word corresponding to each row and each column of the matrix. The information processing apparatus according to claim 12, wherein the information processing apparatus is performed by representing a set of feature words. The feature word generation means further includes, for each predetermined case particle, morpheme data detected between the predetermined case particles when the first part of speech information includes the predetermined case particle. A second feature word corresponding to the first feature word composed of backward morpheme data that is morpheme data detected after the predetermined case particle is generated based on the second rule,
The information processing apparatus according to claim 8, wherein the output unit outputs the second feature word corresponding to the first feature word together with the first feature word. The second rule is that, of the backward morpheme data for each predetermined case particle, backward morpheme data whose part of speech belongs to a part of speech other than the second category is determined from the backward morpheme data detected immediately after the predetermined case particle. The information processing apparatus according to claim 14, wherein the backward morpheme data until detection is combined in the order of detection. The information processing apparatus according to claim 15, wherein the second classification is a noun and a verb. The document data is a plurality of document data,
The output means further includes an appearance frequency calculating means for calculating an appearance frequency in the plurality of document data for each front morpheme data of the plurality of document data.
The appearance frequency of the morpheme data constituting the first feature word is an appearance frequency calculated by the appearance frequency calculation means of each front morpheme data constituting the first feature word,
The output of the first feature word of each document data and the second feature word corresponding to the first feature word is performed in the order of the appearance frequency of the front morpheme data of each first feature word. The information processing apparatus according to claim 14. The information processing apparatus further includes:
Of the front morpheme data of the plurality of document data, for the front morpheme data constituting the first feature word, the appearance frequency in each document data is calculated, and a document vector having the calculated appearance frequency value as an element is calculated. A similarity calculating means for generating and calculating a similarity between each of the plurality of document data using each document vector;
For each similarity calculated by the similarity calculation means, a combination having the maximum similarity is extracted, and a set of each reference link is generated using each document data included in each extracted set as each reference document data. The connection direction of one of the two connection directions starting from the reference document data of the reference connection is set as one reference connection of the reference connection set, and the other connection direction is set as the other reference connection of the set. Set, sequentially select a set of document data in descending order of similarity, perform a connection process to connect each selected set of document data to each reference connection according to a predetermined condition, Determining means for determining the output order of the document data;
The predetermined condition includes a first condition and a second condition, and the first condition is that, for each document data of each combination, the one document data is already connected to any one of the standard connections. The second condition is that any document data is unlinked,
The concatenation process is performed when the document data of any one of the selected combinations satisfies the first condition, and the combination is set in the concatenation direction set in the standard concatenation including the connected document data satisfying the first condition. When both document data of the selected combination and the selected combination satisfy the second condition, a new reference connection is made using both document data as new reference document data. A second connection process for generating a set of the above, and connecting the reference links of each set according to the number of each document data connected to each of the reference links, the reference link of each connected group based on a predetermined rule Including a third connection process to be connected,
The output means outputs the first feature word and the second feature word of each document data for each document data according to the order of each document data determined by the determination means. The information processing apparatus according to 17. A document analysis method for performing morphological analysis processing on document data, detecting morphemes in the document data, decomposing the document data into morpheme data, and analyzing the document data,
A feature word generation step of generating a first feature word composed of the morpheme data based on a predetermined first rule;
An output step of determining an output rank of the first feature word based on an appearance frequency of the morpheme data constituting the first feature word, and outputting the first feature word according to the output rank. Characteristic document analysis method. A program to be executed by an information processing apparatus including a CPU,
Performing morpheme analysis on the document data, detecting morphemes in the document data, and decomposing the document data into morpheme data;
A feature word generation step of generating a first feature word composed of the morpheme data based on a predetermined first rule;
An output step of determining an output rank of the first feature word based on an appearance frequency of the morpheme data constituting the first feature word, and outputting the first feature word according to the output rank. A featured program. The storage means stores a plurality of the patent document data,
The feature word generation means uses the morpheme data of a first predetermined portion including a character string indicating a technical feature constituting the invention of each patent document data in the claim data of each patent document data. A third feature is generated by generating the first feature word and using the morpheme data of a second predetermined portion including a character string indicating an object of invention of the patent document data in the claim data of each patent document data. Generate words,
The information processing apparatus further includes:
The plurality of patent document data is clustered using first appearance frequencies in the plurality of patent document data of the morpheme data included in the third feature words, and the patent documents corresponding to the third feature words Cluster identification means for identifying the cluster to which the data belongs;
A document vector of each patent document data is generated based on a second appearance frequency in the plurality of patent document data of each first feature word, and each first feature word is defined as an observation variable using each document vector. Factor analysis means for performing factor analysis to calculate the factor loading of each first feature word and the factor score of each patent document data;
Factor identifying means for identifying the factor of each first feature word based on the factor loading, and for identifying the factor of each patent document data based on the factor score;
A technical element keyword indicating the factor is generated using the first feature word corresponding to each factor identified by the factor identifying unit, and the patent document data belonging to each cluster identified by the cluster identifying unit Keyword generating means for generating a product group keyword indicating the cluster using the third feature word;
The output means, as information representing the tendency of the plurality of patent document data, based on the factor of each patent document data specified by the factor specifying means, the relationship between each technical element keyword and each product group keyword The information processing apparatus according to claim 7, wherein the relationship information is output. The claim data includes claim data for each claim,
When generating the first feature word, the feature word generation means uses the morpheme data of the first predetermined portion of each claim data in the claim data of the patent document data, and 22. The morpheme data of the second predetermined portion of the predetermined claim data in the claim data of each patent document data is used when generating three characteristic words. Information processing device. The factor specifying means specifies, as the factor of the first feature word, a factor for which the factor loading amount of each first feature word calculated by the factor analyzing means is equal to or greater than a first threshold, and the factor analyzing means The information processing apparatus according to claim 21, wherein a factor having the calculated factor score of each patent document data equal to or greater than a second threshold is specified as a factor of the patent document data. The clustering by the cluster specifying means generates a document vector of each third feature word based on a third appearance frequency in each third feature word of each morpheme data of the second predetermined portion, and each third Calculating a similarity between the document vectors of the third feature word having a fourth appearance frequency of the feature word in the plurality of patent document data equal to or higher than a predetermined value, and extracting a cluster according to the similarity; A process of calculating a similarity between the third feature word and the cluster having a fourth appearance frequency smaller than the predetermined value, and including the patent document data of the third feature word in the cluster according to the similarity The information processing apparatus according to claim 21, further comprising: The keyword generating unit combines the first feature words having the factor loading amount of the factor equal to or greater than a third threshold among the first feature words corresponding to the factors specified by the factor specifying unit. For each cluster extracted by the cluster specifying means, the technical element keyword is generated, and the similarity between the centroid vector of the cluster and the document vector of the third feature word of the patent document data belonging to the cluster is calculated. The information processing apparatus according to claim 21, wherein the product group keyword is generated by calculating and combining the third feature words of the patent document data belonging to the cluster according to the similarity. The output means counts, for each product group keyword, the number of cases for each factor of the patent document data belonging to the cluster corresponding to the product group keyword, and uses the factor of each product group keyword as the relation information. The information processing apparatus according to claim 21, wherein information associating the number of cases with a technical element keyword indicating the factor is output. The storage means further stores evaluation values corresponding to the plurality of patent document data,
The output means, for each product group keyword, aggregates the evaluation values of the patent document data belonging to the cluster corresponding to the product group keyword for each factor, and uses the product group keyword as the relation information. The information processing apparatus according to claim 21, further comprising: outputting information in which evaluation results for each factor are associated with a technical element keyword indicating the factor.

Images