JP5292873B2 - Document search method, document search device, and recording medium recording document search program - Google Patents

Description

  The present invention relates to a document search method, a document search apparatus, and a recording medium on which a document search program is recorded for searching for documents to which classification is assigned.

  Documents such as patents and papers have document information, and hierarchical classification is given to the document information. When searching such a document by a keyword, the following two cases may occur.

(Case 1) The same word has a plurality of meanings.
(Case 2) The same concept is expressed by different words.

  Case 1 may cause search noise (unintended documents are included in the search results), and case 2 may cause search leakage (target documents are not included in the search results).

  As a technology to solve such problems (search noise and search more), the user selects the appropriate category or facet, and from the selected category or facet, the concept name organized in that category or facet Patent classification information search processing and a patent classification information search program storage medium (Patent Document 1) using a classification code for narrowing down classification items based on this are known.

Patent documents include IPC (International Patent Classification), Japanese FI (File Index) and F (File Forming) terms, and US patent classification (US patent classification) types of hierarchical classification codes. It is given in the literature. In particular, in FI and F-term, the classification code values are subdivided into 100,000 items or more. Therefore, when the user can know the value of the classification code suitable for the purpose and purpose of the search, the search is performed by using the technique of Patent Document 1 using the value of the classification code as a key. It is possible to realize a highly accurate search with less noise and search leakage.
By the way, although different from the above-mentioned patent classification, various types of documents such as papers and books are also defined and assigned by information providers such as standard classification codes such as International Decimal Classification (UDC) and JST classification. There are various classification codes.
JP 11-328192 A

  However, not only patent documents such as those described above, but in searching for various documents, if not only Japanese documents but also US documents are searched, the searcher may not be familiar with the classification of the United States, or appropriate English It is difficult to derive an appropriate classification from the search keyword if the technical term (translation) is not known.

  In addition, when examining the US classification based on Japanese keywords, the translation results for the Japanese keywords and the ambiguity of the US classification mixed the search results with literature that was noisy (related but different subject matter). As a result, the accuracy of the candidate list for the document classification code may be reduced. Therefore, it is necessary to verify whether or not the search result is appropriate.

  The present invention has been made to solve the above-described problems, and uses a search result obtained by searching for a first document group composed of a plurality of documents. The purpose is to search for documents included in the second document group.

  In order to solve the above-described problem, a document search method uses a search result obtained by searching for a first document group including a plurality of documents, to obtain a second document group including a plurality of documents. A method for searching for an included document, wherein a computer accepts a specific word or phrase or a specific classification item for classifying a document as a search condition, and the first document based on the search condition A first search step for searching for a predetermined document included in the group, and related information for associating the first document group with the second document group included in the document searched in the first search step. The second document group is searched using the related information as a search condition, the classification item given to the document extracted by the search is specified, and the second document group is again determined based on the specified classification item. search for A second search step, and a third search step of specifying the related information included in the document searched in the second search step, and searching the first document group using the related information as a search condition; The phrase or classification item included in the document searched from the first document group in the third search step is specified, and the degree of coincidence with the word or classification item received as the search condition in the search reception step is calculated And an output step for outputting search result information indicating the degree of coincidence.

  In addition, the document search apparatus searches for a document included in the second document group including a plurality of documents by using a search result obtained by searching the first document group including a plurality of documents. A device that accepts a specific word or phrase or a specific classification item for classifying a document as a search condition, and searches for a predetermined document included in the first document group based on the search condition First retrieval means, and related information relating to the first document group and the second document group included in the document retrieved by the first retrieval means, and using the relevant information as a search condition Second search means for searching the second document group, specifying a classification item given to the document extracted by the search, and searching the second document group again based on the specified classification item; , Searched by the second search means A third search means for searching for the first document group using the related information as a search condition, and within the first document group searched by the third search means. Output means for specifying a phrase or classification item included in the document, calculating a matching degree with the phrase or classification item received by the search receiving means, and outputting search result information indicating the matching degree.

  In addition, the document search program stored in the storage medium may be configured such that the computer accepts a specific word / phrase or a specific classification item for classifying the document as a search condition, and the first search based on the search condition. First search means for searching for a predetermined document included in the document group, and related information relating the first document group and the second document group included in the document searched by the first search means A second search means for searching the second document group using the related information as a search condition, specifying a classification item assigned to the document extracted by the search, and the second search Specifying the related information contained in the document searched by the means, and using the related information as a search condition, a third search means for searching the first document group, and the search performed by the third search means First Function as output means for specifying a phrase or classification item included in a document in a book group, calculating a degree of coincidence with the word or classification item received by the search means, and outputting search result information indicating the degree of coincidence Let

  According to the present invention, a document included in a second document group composed of a plurality of documents is retrieved using a search result obtained by searching the first document group composed of a plurality of documents. Can do.

  Embodiments of the present invention will be described below with reference to the drawings.

Embodiment 1 FIG.

In the present embodiment, a description will be given by taking as an example a patent document search apparatus that searches for a classification code assigned to a document, with the document being a patent publication. In addition, this patent document search device searches a USC or IPC classification that matches a keyword or patent classification based on a Japanese keyword or a patent classification effective in Japan, but is merely an example. It is not limited by country (Japan and US) and patent classification.
Further, the present invention is not limited to the patent document search apparatus, and can be realized in the same manner as the present embodiment as long as it is various document search apparatuses that use a classification system for classifying documents.

  First, the configuration of the document search apparatus according to this embodiment will be described.

  FIG. 1 is a block diagram showing an example of the configuration of the patent document search apparatus according to the first embodiment. This patent document search apparatus includes a control unit 1 (extraction unit, coincidence calculation unit, association unit) that performs search control, and a search unit 2 (first search unit, second search unit, first search unit) that searches patent documents. 3 search unit), an input unit 3 (key information acquisition unit) for receiving input key information from the user, a display unit 4 for displaying to the user, a storage unit 5 for storing input key information, search results, and the like, Japan ( A Japanese patent document DB (Database) 6 that manages patent documents in the first country) and a US patent document DB 7 that manages patent documents in the United States (second country) are provided.

  Next, the operation of the patent document search apparatus according to this embodiment will be described. FIG. 2 is a flowchart showing the operation of the patent document search apparatus according to the first embodiment. FIG. 3 is a diagram illustrating an example of classification code extraction. FIG. 4 is a diagram showing an example of classification code search. FIG. 5 is a diagram illustrating an example of keyword extraction and aggregation when keywords are input. FIG. 6 is a diagram illustrating an example of keyword extraction and aggregation when a classification code is input. Hereinafter, the flowchart of FIG. 2 will be described with reference to FIGS. In the following description, the search unit searches for a Japanese patent document DB when searching for a Japanese patent document (first country patent document), and a US patent when searching for a US patent document (second country patent document). Assume that each document DB is accessed.

  First, the input unit 3 receives input key information (keywords, classification codes such as IPC, FI, and F terms, key information) from the user and passes it to the control unit 1, and the control unit 1 stores the input key information. The search unit 2 instructs the search unit 2 to execute a US patent document search process, which will be described later, and the search unit 2 executes the US patent document search process (S101, first search step, second search step) ). This US patent document search process is a process of searching for US patent documents from input key information that is a Japanese keyword or a classification code other than USC. And

  When the US patent document set is obtained, the control unit 1 extracts the classification code attached to each patent document in the US patent document set as shown in FIG. 3 (S102, second search step). In the extraction of the classification code, the control unit 1 extracts one classification code as a classification code of a plurality of hierarchies in consideration of the hierarchy of the classification code. For example, if the classification code is IPC, there is a hierarchy of sections, classes, subclasses, main groups, and subgroups from the upper level, the higher level is a rough classification, and the lower level is a finer classification. Taking FIG. 3 as an example, when B25J5 / 00, which is an IPC classification code attached to a US patent document, is extracted, this classification code is B25J (subclass), B25J5 / (main group), B25J5 / 00 (sub Each group is extracted as a classification code.

  In this way, searching the US patent document DB 7 again by using the classification code attached to each patent document in the US patent document set can prevent search leakage. Further, a wider range of patent documents can be retrieved by searching the US patent document DB 7 using the higher-level classification codes of the obtained classification codes.

  When a classification code is extracted from a set of US patent documents, the control unit 1 aggregates and ranks the extracted classification code candidates, and creates a classification code list that is a ranked list of the extracted classification codes (S103, Second search step). In this classification code list, for example, classification codes are arranged in descending order of the number of cases, such as 100 cases B25J, 35 cases B25J5 / 10, A63H.

  Next, the control unit 1 sequentially inputs each classification code in the classification code list. At this time, the control unit 1 determines whether there is a classification code to be input (S104).

  When there is a classification code to be input (S104, YES), the control unit 1 instructs the search unit 2 to search for a US patent document including the input classification code. As shown in FIG. 5, a US patent document is searched based on the input classification code using the index file (index table) of the US patent document DB 7 (S105, second search step). Note that the US patent documents searched here are collectively referred to as a second US patent document set.

  Next, the control unit 1 instructs the search unit 2 to execute a Japanese patent document search process to be described later, and the search unit 2 executes the Japanese patent document search process (S106). This Japanese patent document search process is a process of searching for a second Japanese patent document set that is a Japanese patent document corresponding to the second US patent document set.

  When the Japanese patent document search process is executed by the search unit 2 and the second Japanese patent document set corresponding to the second US patent document set is searched, the control unit 1 displays the case shown in FIG. Example), or extraction of keys (keywords or classification codes) from each Japanese patent document of the second Japanese patent document set, as shown in FIG. 6 (example when the user input is a classification code), Aggregation and ranking are performed to create a key information list that is a list with key rankings (S107, extraction step). In the key information list, when the input by the user is a keyword, the keywords are arranged in descending order, such as 100 robots, 35 walking, 10 toys, and so on. On the other hand, when the input by the user is a classification code, the key information list is a list of classification codes arranged in descending order, such as 100 cases B25J5 / 00, 35 cases A63H11 / 18, 10 cases B62D52 / 02. It becomes. Also, this key information list is assumed to be associated with the USC (or IPC) classification code of the US patent document corresponding to the Japanese patent document of the second Japanese patent document set.

  Next, the control part 1 performs the suitability determination process mentioned later (S108, a coincidence degree calculation step, an association step). This suitability determination process is a process for comparing the input key information stored in the storage unit 5 with the key information list to determine the suitability of the classification code.

  When the suitability determination process determines the suitability of the classification codes in the classification code list, the control unit 1 determines whether there is a classification code to be input next in the classification code list (S104). .

In this determination, when there is no classification code to be input next (S104, NO)
The control unit 1 executes suitability determination result display processing described later (S109). This suitability determination result display process is a process for causing the display unit 4 to display suitability for each classification code in the classification code list determined by the suitability determination process.

  Based on the keyword input by the user, FI or F-term that is a classification code unique to Japan, or an IPC classification code that is an international patent classification (these classifications are set as the first classification) by the above-described operation, The USC or IPC classification codes corresponding to these US patent classifications can be searched. Hereinafter, operations of the US patent document search process, the Japanese patent search process, the suitability determination process, and the suitability determination result display process in the flowchart of FIG. 2 will be described.

  First, the US patent document search process that is the operation of step S101 in the flowchart of FIG. 2 will be described. FIG. 7 is a flowchart showing the operation of the US patent document search process. Moreover, FIG. 8 is a figure which shows an example of the search of the Japanese patent document by a keyword. FIG. 9 is a diagram showing an example of a search for Japanese patent documents using classification codes. FIG. 10 is a diagram illustrating an example of extraction of citation information. Hereinafter, the flowchart of FIG. 7 will be described with reference to FIGS.

  First, when the search unit 2 receives an instruction for US patent document search processing from the control unit 1, as shown in FIG. 8 (when the user inputs a keyword) or FIG. 9 (when the user inputs a classification code), as shown in FIG. In addition, based on the input key information (keyword or classification code) that is the key input to the input unit 3 by the user, the Japanese Patent Literature DB is searched (S201, first search step). The Japanese patent documents searched in this search are set as the first Japanese patent document set.

  When the Japanese patent document is searched by the search unit 2, the control unit 1 is the patent number of the US patent document cited in the Japanese patent document from each Japanese patent document in the first Japanese patent document set. First citation information (first related information) is extracted (S202, second search step).

  Further, the search unit 2 searches the US patent document DB 7 based on the extracted first citation information (S203, second related step). The US patent document that is the search result is defined as a set of US patent documents.

  By this US patent document search process, a US patent document corresponding to a Japanese patent document can be searched.

  Next, the Japanese Patent Literature search process that is the operation of Step S106 in the flowchart of FIG. 2 will be described. FIG. 11 is a flowchart showing the operation of the Japanese patent document search process. FIG. 12 is a diagram illustrating an example of a Japanese patent document search process.

  First, as shown in FIG. 12, the control unit 1 uses the second citation information (second related information) that is the patent number of the Japanese patent document cited in each US patent document of the second US patent document set. Is extracted (S301, third search step).

  Further, the control unit 1 searches the Japanese patent document DB 6 based on the extracted second citation information (S302, third search step). The Japanese patent document that is the search result is set as the second Japanese patent set.

  By this Japanese patent document search process, a Japanese patent document corresponding to the US patent document can be searched.

  Next, the suitability determination process that is the operation of step S108 in FIG. 2 will be described. FIG. 13 is a flowchart showing the operation of the suitability determination process. FIG. 14 is a diagram illustrating an example of matching between input key information and extracted key information. FIG. 15 is a diagram illustrating an example of scoring of extracted key information.

  First, the control unit 1 reads the input key information stored in the storage unit 5, and a word frequently appearing in a certain number or more in each Japanese patent document in the second Japanese patent document set or each Japanese patent in the second Japanese patent document. The classification code assigned to the document is extracted as extracted key information, and the input key information and the extracted key information are compared by two methods of complete matching and partial matching (S401, matching degree calculation step). Specifically, as shown in FIG. 14, for example, when the input key information is a keyword “speech recognition”, the control unit 1 divides this input key information into “speech” and “recognition”. “Speech” and “recognition” are partially matched, and “speech recognition” and “speech recognition device” are completely matched. If the input key information has a classification code of “B25J5 / 00”, the input key information is divided into hierarchies such as “B25J”, “B25J5”,. And B25J5 / 00 is completely coincident. This extraction key is assumed to be associated with the classification code of the US patent document corresponding to the Japanese patent document as the extraction source.

  Further, the control unit 1 calculates a score based on the presence / absence of extracted key information of complete matching and partial matching of keywords or classification codes in the key information list (S402, matching degree calculation step, association step). For example, as shown in FIG. 15, 1 point is obtained for partially matched extracted key information, 2 points are obtained for completely matched extracted key information, and the other extracted key information is 0 points, corresponding to 704/275. All the scores of the extracted key information are added to obtain a score of 704/275. Further, without considering the number of partial matches and complete matches, 2 points may be used when the extracted key information includes a complete match, and 1 point may be used when only the partial match is included. Considering the ranking in the extracted key information, for example, if the first keyword is partially matched or completely matched, the point for that keyword is 10 times. If the second keyword is partially matched or completely matched, the point for that keyword is The number of points may be changed according to the ranking, such as 9 times. These score calculation methods may be combined.

  With this suitability determination process, the suitability of the classification code corresponding to the input key information can be determined.

  Next, the suitability determination result display process showing the operation of step S109 in the flowchart of FIG. 2 will be described. FIG. 16 is a diagram illustrating a result display screen.

When the suitability determination process is performed for all the classification codes in the key information list, the controller 1 determines the suitability for each USC that is a classification code of the US patent document, as shown in FIG. A search result screen showing the degree of coincidence of the results is displayed. In this display, each row represents a classification code, and each column represents ranking, hit count, total count, relevance, relative score, USC, and related key.
In the example of FIG. 16, they are arranged in descending order of the degree of relevance with respect to the input key information that is a search condition. It is displayed by changing the character color of USC depending on the height.

  The relevance is the product of the appearance frequency of the input key information in all documents and the appearance frequency of each classification code in the key information list, and the largest relevance value is 1.00. The rank is a rank assigned from a classification code having a high degree of association. The number of hits is displayed in the form of (number of occurrences of classification code) / (number of documents hit by input key information). In other words, (the number of documents in which the input key information is hit) is the number of search results of US patent documents derived from related information included in Japanese patent documents, and (the number of classification codes appears) is This is the number of cases where USC, which is the classification code of each row, appears in the number of search results of US patent documents. The total number of documents is the total number of documents having a classification code in each row among all US patent documents. The relative score is a bar graph obtained by converting the degree of association into a relative value with respect to the maximum value. USC is a classification code of the US patent classification included in the key information list. Further, the related key is a word that appears more than a certain number in Japanese patent documents corresponding to US patent documents including the classification code of each row, and is extracted key information associated with the classification code of each row. This related key is displayed when the input key information is a keyword, and when the input key information is a classification code, a classification code is displayed.

  Each extracted key in the related key is displayed in a different color depending on the determination in the suitability determination process described above. For example, with respect to the input key information, an extraction key with a perfect match is displayed in red, an extraction key with a partial match is displayed in orange, and an extraction key with a mismatch is displayed in black. Similarly, USC also displays different colors depending on the score of the corresponding key. For example, if the related key score is 3 points or more, it is displayed in red, if the related key score is 2 points, it is displayed in orange, and if the related key score is 1 point, it is displayed in yellow. When the score of the related key is 0 point, it is displayed in black.

  With the above configuration and operation, it is possible to search for a US patent classification (or IPC classification) that matches the input, using Japanese keywords or Japanese patent classification (including IPC classification) as input.

Embodiment 2. FIG.

  In the first embodiment, the US patent document is searched using the cited information of the Japanese patent document searched based on the input key information, and the corresponding Japanese patent document is searched using the cited information of the US patent document. However, the present embodiment is different from the first embodiment in that PCT application information is used instead of citation information. Hereinafter, operations different from those of the first embodiment will be described.

  The patent document search apparatus according to the present embodiment is different from the patent document search apparatus according to the first embodiment in the operations of the US patent document search process and the Japanese patent document search process. First, the operation of the Japanese patent document search process according to the present embodiment will be described with reference to the following diagram. FIG. 17 is a flowchart showing the operation of the Japanese patent document search process according to the second embodiment. FIG. 18 is a diagram illustrating an example of extraction of PCT application information.

  First, when the search unit 2 receives an instruction for US patent document search processing from the control unit 1, as shown in FIG. 8 (when the user inputs a keyword) or FIG. 9 (when the user inputs a classification code), as shown in FIG. In addition, based on the input key information (keyword or classification code) that is the key input to the input unit 3 by the user, the Japanese Patent Literature DB is searched (S501, first search step). The Japanese patent documents searched in this search are set as the first Japanese patent document set.

  When the Japanese patent document is searched by the search unit 2, the control unit 1, as shown in FIG. 18, the US patent document corresponding to the PCT application in the first Japanese patent document set or the Japanese patent document as the priority claim. The first PCT application information (first related information), which is information indicating the above, is extracted (S502, second search step).

  Further, the search unit 2 searches the US patent document DB 7 based on the extracted PCT application information (S503, second search step). The US patent document that is the search result is defined as a set of US patent documents.

  Next, the operation of the US patent document search process according to the present embodiment will be described with reference to the following diagram. FIG. 19 is a flowchart showing the operation of the Japanese patent document search process according to the second embodiment. FIG. 20 is a diagram illustrating an example of extraction of PCT application information.

  First, as shown in FIG. 19, the control unit 1 includes second PCT application information (first PCT application information (first 2 related information) is extracted (S501, third search step).

  Further, the control unit 1 searches the Japanese Patent Literature DB 6 based on the extracted second PCT application information (S502, third search step). The Japanese patent document that is the search result is set as the second Japanese patent set.

  As described above, the patent document search apparatus according to the present embodiment can search for a patent document corresponding to a PCT application or a patent document as a priority claim, thereby searching for a patent document in another country that matches. It is possible to search for a patent document that is more suitable than the cited document.

Embodiment 3 FIG.

  In the second embodiment, PCT application information is used instead of the citation information. However, this embodiment is different from the second embodiment in that the IPC classification is used instead of the citation information. Hereinafter, operations different from those of the second embodiment will be described.

  The patent document search apparatus according to the present embodiment is different from the patent document search apparatus according to the second embodiment in the operations of the US patent document search process and the Japanese patent document search process. First, the operation of the Japanese patent document search process according to the present embodiment will be described with reference to the following diagram. FIG. 21 is a flowchart showing the operation of the Japanese patent document search process according to the third embodiment. FIG. 22 is a diagram illustrating an example of IPC classification extraction.

  First, when the search unit 2 receives an instruction for US patent document search processing from the control unit 1, as shown in FIG. 8 (when the user inputs a keyword) or FIG. 9 (when the user inputs a classification code), as shown in FIG. In addition, based on input key information (keyword or classification code) that is a key input to the input unit 3 by the user, the Japanese Patent Literature DB is searched (S701, first search step). The Japanese patent documents searched in this search are set as the first Japanese patent document set.

  When the search unit 2 searches for Japanese patent documents, the control unit 1 extracts the IPC classification (first related information) in the first Japanese patent document set as shown in FIG. 22 (S702, second search step). .

  Further, the search unit 2 searches the US patent document DB 7 based on the extracted IPC classification (S703, second search step). The US patent document that is the search result is defined as a set of US patent documents.

  Next, the operation of the US patent document search process according to the present embodiment will be described with reference to the following diagram. FIG. 23 is a flowchart showing the operation of the Japanese patent document search process according to the third embodiment. FIG. 24 is a diagram illustrating an example of IPC classification extraction.

  First, as shown in FIG. 24, the control unit 1 extracts the IPC classification (second related information) in the second US patent document set (S801, third search step).

  Further, the control unit 1 searches the Japanese Patent Literature DB 6 based on the extracted IPC classification (S802, third search step). The Japanese patent document that is the search result is set as the second Japanese patent set.

As described above, the patent document search apparatus according to the present embodiment can search for patent documents in other countries that are compatible by using the IPC classification that is an internationally defined classification, and at least in the IPC classification. Can search patent documents with the same classification. Further, for the extracted IPC classification, for example, a low weight is given to the IPC classification given to many patents, and a high weight is given to the IPC classification given to a small number of patents. As described above, weighting is performed based on the frequency, and only the IPC classification with a certain weight or more may be used for the search.

Embodiment 4 FIG.
In the above-described embodiment, citation information, PCT application information, and IPC classification are used as information (first related information and second related information) for searching US patent documents from Japanese patent documents. In the embodiment, information (first related information) for searching US patent documents from Japanese patent documents and information (second related information) for searching Japanese patent documents from US patent documents are different information. Can be selected. FIG. 25 is a diagram illustrating an example of a condition setting screen.

  In inputting key information for search, the control unit 1 causes the display unit 4 to display a condition setting screen as shown in FIG. The user can select information for searching for patent documents and information for searching Japanese patent documents from US patent documents. In FIG. 25, the primary search indicates the conditions for the US patent document search process, and the secondary search indicates the conditions for the Japanese patent document search process. When the conditions are set, the patent document search apparatus according to the present embodiment performs the US patent document search process and the Japanese patent document search described in the first, second, and third embodiments according to the setting. Each process is executed.

As described above, in the US patent document search process and the Japanese patent document search process, it is possible to adjust the width of documents to be searched by enabling selection of information used for the search. For example, when the IPC classification is used in the US patent document search process, a document having a similar IPC classification is searched, and therefore, a US patent document with high accuracy can be searched for the input key information. Further, when citation information is used in the Japanese patent document search process, it is possible to search for US patent documents having similar contents even in different classifications.

  Further, the document retrieval apparatus according to the present embodiment can be easily applied to the information processing apparatus, and can further improve the performance of the information processing apparatus. Here, the information processing apparatus may include, for example, a server, a PC (Personal Computer), and the like.

  Furthermore, a program for executing the above steps in a computer constituting the patent document search device can be provided as a patent document search program. By storing the above-described program in a computer-readable recording medium, the computer constituting the patent document search apparatus can be executed. Here, examples of the recording medium readable by the computer include an internal storage device such as a ROM and a RAM, a portable storage such as a CD-ROM, a flexible disk, a DVD disk, a magneto-optical disk, and an IC card. It includes a medium, a database holding a computer program, another computer and its database, and a transmission medium on a line.

(Additional remark 1) The document search method which searches the document contained in the 2nd document group comprised from a some document using the search result searched for the 1st document group comprised from the some document as object Because
Computer
A search accepting step for accepting a specific classification item for classifying a specific word or phrase or document as a search condition;
A first search step of searching for a predetermined document included in the first document group based on the search condition;
The related information that associates the first document group with the second document group included in the document searched in the first search step is specified, and the second document group is searched using the related information as a search condition. A second search step of specifying a classification item given to the document extracted by the search, and searching the second document group again based on the specified classification item;
A third search step of identifying the related information included in the document searched in the second search step, and searching the first document group using the related information as a search condition;
The phrase or classification item included in the document searched from the first document group in the third search step is specified, and the degree of coincidence with the phrase or classification item received as the search condition in the search reception step is calculated. An output step for outputting search result information indicating the degree of coincidence;
A document search method characterized by comprising:
(Supplementary note 2) The document retrieval method according to supplementary note 1, wherein
In the second search step or the third search step, document identification information common to the first document group and the second document group is specified as related information.
(Supplementary note 3) The document retrieval method according to supplementary note 1, wherein
In the second search step or the third search step, a classification item common to the first document group and the second document group is specified as related information.
(Supplementary note 4) The document retrieval method according to supplementary note 1, wherein
In the second search step or the third search step, as the related information, the other cited in the document searched from either one of the first document group or the second document group. A document search method characterized by specifying a document in a document group.
(Supplementary note 5) The document search method according to any one of supplementary notes 1 to 4,
In the second search step, the second document group is searched based on the related information, and the document extracted by the search is used in the first document group given to the extracted document. A document search method characterized by specifying a classification item of a classification system different from the classification system, and searching the second document group again based on the specified classification item.
(Supplementary note 6) The document search method according to supplementary note 1, wherein
In the second search step, a plurality of the classification items assigned to the extracted document are specified, and the second document group is searched again using each of the specified classification items as a search condition,
In the third search step, for each of the identified classification items, the first document group is searched using the related information included in the searched document as a search condition,
In the output step, for each identified category item, a phrase or category item included in a document searched from the first document group is specified, and the phrase or phrase received as a search condition in the search reception step A document search method comprising: calculating a degree of coincidence with a classification item and outputting a search result indicating the degree of coincidence for each classification item specified in the second search step.
(Supplementary note 7) The document search method according to supplementary note 2, wherein
The documents included in the first and second document groups are patent documents,
The document search method, wherein the document identification information common to the first document group and the second document group is PCT application information of patent documents.
(Supplementary note 8) The document retrieval method according to supplementary note 3, wherein
The documents included in the first and second document groups are patent documents,
The document retrieval method according to claim 1, wherein the classification items common to the first document group and the second document group are classification items based on an IPC classification system for classifying patent documents.
(Supplementary note 9) The document search method according to supplementary note 4, wherein
The documents included in the first and second document groups are patent documents,
A document search characterized by specifying an identification number of a patent document in the other document group that is cited in the patent document searched from either one of the first or second document group Method.
(Additional remark 10) The document search apparatus which searches the document contained in the 2nd document group comprised from a some document using the search result searched for the 1st document group comprised from the some document as object Because
Search accepting means for accepting a specific term or a specific classification item for classifying a document as a search condition;
First search means for searching for a predetermined document included in the first document group based on the search condition;
The related information relating to the first document group and the second document group included in the document searched by the first search means is specified, and the second document group is selected using the related information as a search condition. A second search means for searching, specifying a classification item given to the document extracted by the search, and searching the second document group again based on the specified classification item;
Third search means for specifying the related information contained in the document searched by the second search means, and searching the first document group using the related information as a search condition;
The phrase or classification item included in the document in the first document group searched by the third search means is specified, the degree of coincidence with the word or classification item received by the search reception means is calculated, and the match A document search apparatus comprising: output means for outputting search result information indicating a degree.
(Supplementary note 11) The document search device according to supplementary note 10, wherein
In the second search means or the third search means, document identification information common to the first document group and the second document group is specified as related information.
(Supplementary note 12) The document search device according to supplementary note 10, wherein
In the second search means or the third search means, a classification item common to the first document group and the second document group is specified as related information.
(Supplementary note 13) The document search device according to supplementary note 10,
In the second search means or the third search means, the other cited in the document searched from either one of the first or second document group as the related information. A document search apparatus characterized by identifying documents in a document group.
(Supplementary note 14) The document search device according to any one of supplementary notes 10 to 13,
In the second search means, the second document group is searched based on the related information, and the document extracted by the search is used in the first document group given to the extracted document. A document search apparatus characterized by specifying a classification item of a classification system different from the classification system and searching the second document group again based on the specified classification item.
(Supplementary note 15) The document search device according to supplementary note 10,
In the second search means, a plurality of the classification items assigned to the extracted document are specified, and the second document group is searched again using each of the specified classification items as a search condition,
In the third search means, for each of the identified classification items, the first document group is searched using the related information included in the searched document as a search condition,
In the output means, for each identified classification item, the phrase or classification item included in the document retrieved from the first document group is identified, and the word or phrase received as a search condition in the search reception means A document search apparatus characterized by calculating a degree of coincidence with a classification item and outputting a search result indicating the degree of coincidence for each classification item specified by the second search means.
(Supplementary note 16) The document search device according to supplementary note 11, wherein
The documents included in the first and second document groups are patent documents,
The document identification apparatus characterized in that the document identification information common to the first document group and the second document group is PCT application information of patent documents.
(Supplementary note 17) The document search device according to supplementary note 12,
The documents included in the first and second document groups are patent documents,
The document retrieval apparatus according to claim 1, wherein the classification items common to the first document group and the second document group are classification items based on an IPC classification system for classifying patent documents.
(Supplementary note 18) The document search device according to supplementary note 13, wherein
The documents included in the first and second document groups are patent documents,
A document search characterized by specifying an identification number of a patent document in the other document group that is cited in the patent document searched from either one of the first or second document group apparatus.
(Supplementary note 19)
Search accepting means for accepting a specific term or a specific classification item for classifying a document as a search condition;
First search means for searching for a predetermined document included in the first document group based on the search condition;
The related information relating to the first document group and the second document group included in the document searched by the first search means is specified, and the second document group is selected using the related information as a search condition. A second search means for searching, identifying a classification item attached to the document extracted by the search,
Third search means for specifying the related information contained in the document searched by the second search means, and searching the first document group using the related information as a search condition;
Specifying a word or category included in a document in the first document group searched by the third search means, calculating a degree of coincidence with the word or category received by the search means, A computer-readable recording medium on which is recorded a document search program that functions as output means for outputting search result information indicating
(Supplementary note 20) A computer-readable recording medium on which the document search program according to supplementary note 19 is recorded,
In the second search means, a plurality of the classification items assigned to the extracted document are specified, and the second document group is searched again using each of the specified classification items as a search condition,
In the third search means, for each of the identified classification items, the first document group is searched using the related information included in the searched document as a search condition,
In the output means, for each identified classification item, the phrase or classification item included in the document retrieved from the first document group is identified, and the word or phrase received as a search condition in the search reception means A computer-readable record in which a document search program is recorded, wherein the degree of coincidence with a classification item is calculated, and a search result indicating the degree of coincidence is output for each classification item specified by the second retrieval unit Medium.

1 is a block diagram illustrating an example of a configuration of a document search device according to Embodiment 1. FIG. 4 is a flowchart showing an operation of the document search device according to the first embodiment. It is a figure which shows an example of classification code extraction. It is a figure which shows an example of a classification code search. It is a figure which shows an example of keyword extraction and totalization at the time of setting a keyword as input. It is a figure which shows an example of keyword extraction and totalization at the time of using a classification code as input. 6 is a flowchart showing an operation of a US patent document search process in the first embodiment. It is a figure which shows an example of the search of the Japanese patent document by a keyword. It is a figure which shows an example of the search of a Japanese patent document by a classification code. It is a figure which shows an example of extraction of quotation information. 4 is a flowchart showing an operation of a Japanese patent document search process in the first embodiment. FIG. 10 is a diagram illustrating an example of a Japanese patent document search process in the first embodiment. It is a flowchart which shows the operation | movement of a compatibility determination process. It is a figure which shows an example of the matching of input key information and extraction key information. It is a figure which shows an example of scoring of extraction key information. It is a figure which shows a result display screen. 10 is a flowchart showing an operation of a Japanese patent document search process according to the second embodiment. It is a figure which shows an example of extraction of PCT application information. 10 is a flowchart showing an operation of a Japanese patent document search process according to the second embodiment. It is a figure which shows an example of extraction of PCT application information. 10 is a flowchart showing an operation of a Japanese patent document search process according to Embodiment 3. It is a figure which shows an example of extraction of an IPC classification. 10 is a flowchart showing an operation of a Japanese patent document search process according to Embodiment 3. It is a figure which shows an example of extraction of an IPC classification. It is a figure which shows the example of a condition setting screen. 1 is a block diagram illustrating an example of a configuration of a document search device according to Embodiment 1. FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Control part, 2 Search part, 3 Input part, 4 Display part, 5 Memory | storage part, 6 Japanese patent document DB, 7 US patent document DB.

Claims (5)

A document search method for searching for a document included in a second document group based on a search condition used when searching for the first document group,
Computer
Search the first document group based on the received search condition,
Extracting the documents of the second document group associated with the retrieved documents of the first document group;
Searching for the second document group based on the attribute information of the extracted document of the second document group, which is compatible attribute information that matches the search condition;
A document search method characterized by the above. The document search method according to claim 1,
2. The document search method according to claim 1, wherein the extraction is to extract a document of the second document group cited in a document of the searched first document group. The document search method according to claim 1,
The first document group and the second document group are document groups according to documents related to patent applications in different countries,
The extraction is to extract a document of the second document group that has a relation by a PCT application with a retrieved document of the first document group or a relation by priority. Document search method. A document search device for searching for a document included in a second document group based on a search condition used when searching for the first document group,
  First search means for searching for the first document group based on an accepted search condition;
  Extracting means for extracting the document of the second document group associated with the searched document of the first document group;
  Second search means for searching the second document group based on the attribute information of the extracted document of the second document group and matching attribute information that matches the search condition;
  A document search apparatus comprising: Computer
  First search means for searching for a first document group based on the received search condition;
  Extracting means for extracting a document of the second document group associated with the searched document of the first document group;
  Second search means for searching the second document group based on the attribute information of the extracted document of the second document group and matching attribute information that matches the search condition;
  A document search program that makes it work.

Images