KR102029916B1 - Intellectual property right document based bigdata analyzing method for design material search - Google Patents

Abstract

The present invention provides a computer implemented method for design material search and analysis based on intellectual property right (IPR) documents, comprising: a step of conducting search in an IPR document database (DB) by using received keywords after receiving the at least two keywords among a material keyword, a technology keyword and an emotion keyword; a step of conducting data mining of documents in a search collection group, which are collected by the search, and acquiring material-technology related IPR documents in which association is evaluated to exist between the material keyword and the technology keyword according to a fixed criterion; a step of conducting opinion mining of the documents in the search collection group, which are collected by the search, and acquiring material-emotion related IPR documents in which association is evaluated to exist between the material keyword and the emotion keyword according to the fixed criterion; a step of acquiring material-technology-emotion related IPR documents corresponding to intersection of the material-technology related IPR documents and the material-emotion related IPR documents; and a step of generating a material-technology-emotion matrix DB in which the acquired material-technology-emotion related IPR documents are structured into a matrix shape according to the material keyword, the technology keyword and the emotion keyword. Therefore, the computer implemented method can be actively utilized in developing a product and a new material for improvement of emotional quality of design.

Description

INTELLECTUAL PROPERTY RIGHT DOCUMENT BASED BIGDATA ANALYZING METHOD FOR DESIGN MATERIAL SEARCH}

The present invention relates to a big data analysis method, and more particularly, to a big data analysis method based on intellectual property (patent / design) document for design material search.

Recently, the importance of emotional quality, such as creative materials, surface treatment, has emerged in the design innovation. This is because design plays an increasing role in shaping creative ideas about the application and application of materials. In this regard, global companies and leading companies are reinforcing their product competitiveness with differentiated strategies such as design, color, material, and method every time they launch new products.

On the other hand, most of SMEs in Korea are having difficulty in designing and manufacturing competitive products that are differentiated from existing products due to lack of information, lack of R & D capability, and lack of awareness of design completion. However, this is not only due to the lack of human, asset and technical capacity of domestic SMEs, but also from the limitations of design methodology and technical expertise.

For example, research on CMF as a conventional design methodology is mostly concentrated in the color area, and thus there is a limit in deriving elements for new materials, methods, and finishing. As one of the methodologies to improve this, there have been efforts to obtain information on new materials, processes, and finishes through advanced intellectual property documents (patents / designs), but this requires skilled search know-how by technical experts or patent experts. In addition, since only the search from a technical point of view is dominant, there has been a difficulty in searching from an emotional quality point of view for the corresponding design element.

As a design element, it is possible to simultaneously improve the emotional quality in the MMF (Material / Methods / Finishing) design process, while the related patent technology and design source data are databased from the point of view of the general researcher or designer, not from the point of view of the patent expert. There is an urgent need for the development of analytical methodologies that can be easily identified and utilized.

The present invention utilizes big data retrieval and analysis based on intellectual property (patent / design) documents to make it easier to obtain material information required for designers or researchers in the industrial / product design process. It is intended to be actively utilized in developing products and new materials for improvement.

According to an aspect of the present invention, as a computer implemented method (Computer implemented method) for searching and analyzing design material based on intellectual property document,

In response to receiving at least two keywords of a material keyword, a technology keyword, and an emotional keyword, performing a search in the intellectual property document DB using the received keyword—wherein the material keyword is a name of a material. The technical keyword is a keyword for expressing a functional characteristic (for example, physical, chemical, biological, electrical, etc. related to physical properties of the material of the material) determined by the material property of the material, The emotional keyword is a keyword expressing an external feature (eg, surface feature, color feature, cognitive / intuitive / emotional feature, etc.) expressed according to the material property of the material;

Performing data mining on documents in the search population collected through the search to obtain material-technical related intellectual property documents which are evaluated to have an association between a material keyword and a technology keyword according to a predetermined criterion;

Performing opinion mining on the documents in the search population collected through the search to obtain material-emotional-related intellectual property documents which are evaluated to have an association between the material keyword and the emotional keyword according to a predetermined criterion;

Obtaining a material-technology-emotional associated intellectual property document corresponding to an intersection of the material-technical intellectual property documents and the material-emotional associated intellectual property documents;

A computer-implemented method is provided that includes generating the material-technology-emotional matrix DB structured in the form of a matrix according to the material keyword, technology keyword, and emotional keyword.

According to an embodiment of the present invention, by using big data retrieval and analysis based on intellectual property (patent / design) document, it is possible to more easily obtain the material information required for designers or researchers in the industrial / product design process In addition, it has an effect that can be actively utilized in the development of products and new materials for improving the emotional quality of design.

In addition, according to the methodology according to the embodiment of the present invention, it is possible to improve the efficiency of new material planning and development that combines technical (functional) elements and emotional elements, and provides a practical search and analysis model for commercialization of material sensitivity technology It can be used for diagnosing emotional quality of target materials and establishing improvement strategies. For companies, it is expected to increase sales of developed products by improving the quality of emotional materials available at the point of contact with consumers.

1 is a view generally showing a big data analysis method based on intellectual property documents for design material search according to an embodiment of the present invention.
2 is a reference diagram for explaining the basic principle of a context window as one of the data mining methodologies introduced in principle in the analysis method of the present invention;
3 is a reference diagram for explaining the basic principle of co-occurrence as one of the data mining methodologies introduced in principle in the analysis method of the present invention.
4 and 5 are reference drawings for explaining the basic principle of Wordnet and its extended version (Senti-Wordnet) as one of opinion mining methodologies introduced in principle in the analysis method of the present invention.
Figure 6 illustrates an embodiment of a method for obtaining mining data and intellectual property documents of a work-technology association according to the analysis method of the present invention.
7 illustrates an embodiment of a method for obtaining mining data and intellectual property documents of a material-emotional association according to the analysis method of the present invention.
8 illustrates an embodiment of a method for obtaining mining data and intellectual property documents of material-technology-emotional association according to the analytical method of the present invention.
9 is an example of a variable matrix structure in which a technical effect is distributed and expressed on a matrix when mining data of a material-technology-emotional relation is matrixed.
10 is an illustration of a valid patent relating to a material obtained according to a technical effect and emotional effect matrix search.
11 to 15 are views illustrating a comparison result between a general patent search method and the method according to the present invention (AMD: Advanced Design Material Discovery).

As the invention allows for various changes and numerous embodiments, particular embodiments will be illustrated in the drawings and described in detail in the written description. However, this is not intended to limit the present invention to specific embodiments, it should be understood to include all transformations, equivalents, and substitutes included in the spirit and scope of the present invention.

In describing the present invention, when it is determined that the detailed description of the related known technology may unnecessarily obscure the subject matter of the present invention, the detailed description thereof will be omitted. In addition, numerals (eg, first, second, etc.) used in the description process of the present specification are merely identification symbols for distinguishing one component from another component. In addition, throughout the specification, when a part is said to "include" a certain component, it means that it may further include other components, without excluding the other components unless otherwise stated.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 is a diagram generally illustrating a big data analysis method based on an intellectual property document for design material search according to an embodiment of the present invention. Referring to Figure 1 describes the overall process of the analysis method according to an embodiment of the present invention.

First, when target keywords (at least two of material keywords, technical keywords, and emotional keywords) to be analyzed are selected, the target keywords are used to perform a search in the intellectual property document DB (Database). Data mining and opinion mining are performed on documents in the search population collected through the search.

That is, in the analysis method of the present invention, data mining is performed to obtain material-technical related intellectual property documents which are evaluated to have an association degree between a material keyword and a technology keyword according to a predetermined criterion, and further, opinion mining is performed to perform a predetermined criterion. Accordingly, obtains material-emotional-related intellectual property documents which are evaluated to have an association between the material keyword and the emotional keyword. In addition, the material-technological-related intellectual property documents obtained according to the intersection of the material-technical-related intellectual property documents and the material-sensitivity related intellectual property documents are obtained. Subsequently, a material-technology-emotion matrix DB is constructed in which the obtained material-technology-emotion-related intellectual property documents are structured in a matrix form according to a material keyword, a technology keyword, and an emotion keyword.

Here, the material keyword is a keyword indicating the name of the material (material), the technical keyword is a functional feature (for example, physical, chemical, related to the physical properties according to the material of the material determined according to the material properties of the material) Biological, electrical characteristics, etc.), and the emotional keyword expresses appearance characteristics (eg, surface characteristics, color characteristics, cognitive / intuitive / emotional characteristics, etc.) expressed according to material properties of the material. Means a keyword.

At this time, in the process of implementing the analysis method of the present invention shown in Figure 1, the data mining / opinion mining methodologies of Figures 2 to 5 were applied in principle.

2 is a reference diagram for explaining the basic principle of the context window as one of the data mining methodologies introduced in principle to the analysis method of the present invention, and FIG. 3 is a principle diagram of the analysis method of the present invention. As a data mining methodology introduced as a reference figure for explaining a basic principle of co-occurrence. The concept of the context window illustrated in FIG. 2 is applied to the principle that a word placed adjacent to a specific reference word is recognized as a high degree of association or recognized as a contextual point of view. In some cases, the contextual relation cannot be secured only by the adjacent operation based on the concept of the context window described above. In this case, it is possible to apply the concept of simultaneous occurrence frequency shown in FIG. Can be. That is, the frequency of co-occurrence is a data mining concept that determines the degree of association between words based on how often the words appear simultaneously or repeatedly.

In the present invention, Wordnet and its extended version (senti-Wordnet) are applied in principle as an opinion mining method. 4 and 5 are reference views for explaining the basic principle of Wordnet and its extended version (Senti-Wordnet) as one of the opinion mining methodologies introduced in principle in the analysis method of the present invention.

WordNet is an English semantic vocabulary building project started at Princeton University in 1985. It is a dictionary of word network structure that combines many English words by synonyms and makes synsets and organizes the relationships among the synsets. SentiWordNet is also a lexical dictionary that scores sentiment scores for WordNet's Synset, which consists of positive (P), negative (N), and neutral (O) and is designed with P + N + O = 1 for one synset. It is. Herein, if P = 1, the synset has an intensely positive meaning, whereas if N = 1, an intensely negative meaning, P = 0 and N = 0 are determined to be neutral. In the present invention, the basic principle of WordNet and its definite version is introduced when DB synonym dictionary of emotional keyword is made into DB. However, the above-described data mining / opinion mining techniques are not applied to the present invention as they are, but it is clearly clarified that they are introduced according to the analysis method according to the present invention with reference to the basic principles of the above techniques.

In the above, the data mining / opion mining methodologies introduced in principle to the present invention have been described. Hereinafter, the analysis methodology of the present invention shown in FIG. 1 will be described in detail with reference to FIGS. 6 to 10. An analysis method according to an embodiment of the present invention to be described below may be processed by a computing device or system including at least one processor and a memory.

As described above, according to the present invention, when the target keywords (at least two or more of material keywords, technology keywords, and emotional keywords) to be analyzed are selected, the target keywords are used in the intellectual property document DB (Database). A search is performed, and data mining and opinion mining are performed on documents in the search population collected through the search.

For example, as illustrated in FIG. 1, if a plastic material is to be durable and has a luxurious or luxurious feeling, the material keyword may be selected as plastic, the technology keyword as durable, and the emotional keyword as luxury or splendor. There will be. In the present specification, for convenience of explanation and understanding of the invention, it is assumed that the material keyword, the technical keyword, and the emotional keyword are all selected (selected). However, in some cases, only two keywords may be selected. .

For example, if you have a durable and high-quality material, you may want to investigate all materials without particular limitations (for example, plastic or metal). In these cases, only technical keywords and emotional keywords It may be determined. In such a case, by inserting all the material keywords included in the keyword dictionary with reference to the synonym dictionary DB regarding the material keyword to be described later and applying the analysis methodology as shown in FIGS. You might want to check what's going on. From a similar point of view, while only the material keyword and the technology keyword or only the material keyword and the emotion keyword are determined, the remaining keywords may be searched by referring to the corresponding keyword dictionary. To further expand this, it would be possible to investigate in which only one keyword of material / technology / sensitivity was determined, but this would greatly increase the research and computational complexity and would not generally match the research intent of the designer or researcher. In this regard, a separate description through the present specification will be omitted.

As described above, when the material / technology / sensitivity keyword is determined, a search for the intellectual property document and a data mining and opinion mining process to be described later are performed. In this case, the keyword dictionary DB may be used. Of course, the search and mining of the intellectual property document may be performed using only the keywords (words) determined (ie, directly selected by the designer or the researcher). Hereinafter, the keyword expansion search using the keyword dictionary DB will be described.

According to an embodiment of the present invention, a synonym dictionary DB (database) arranged to group the synonyms associated with the corresponding keyword by the material keyword, the technical keyword, and the emotional keyword is managed by representing a specific keyword. And mining processes. Accordingly, in the process of searching for the intellectual property document using the keyword, the extended search and mining including the synonyms of the corresponding keyword may be performed according to the thesaurus dictionary DB.

For example, when "durability" is selected as the technical keyword as in the previous example, various synonyms expressing "durability" differently include "high rigidity", "hard", "hard", "hard", "strong" Etc. can be grouped and managed DB. As another example, when "soft" is selected as an emotional keyword, as various synonyms expressing this differently, "soft", "flexible", "smooth", "soft", "smooth", etc., may be grouped together and managed by DB. It is. In this case, which word to select a representative keyword may be directly selected by the system designer or may be automatically selected according to statistical representativeness of the word or idiomaticity as a technical term. Shall be.

Further, according to another embodiment of the present invention, when any one of the material keywords in the synonym dictionary DB for the material keyword includes a material keyword of a plurality of sub-concepts belonging to the keyword as a higher-level keyword, in the dictionary DB In the upper-level keyword and the lower-level keyword can be linked management. Accordingly, in the process of searching for the intellectual property document using the keyword, the extended search and mining including the material keyword of the subordinate concept managed in association with the corresponding material keyword can be performed.

For example, the material keyword "plastic" is a higher concept that includes the material keywords "PC", "PET", "PP", "PPE", "PA", etc. of the subordinate concept. Therefore, in the search and mining process of the material keyword of the upper concept, the material keyword of the lower concept associated with it may be expanded and searched and mined. To further expand this concept, when "PC" is selected as the material keyword, the user is presented with a "expansion of associated material" button through the user interface screen, and the upper concept material keyword "plastic" or its alternative concept. It is also possible to design the system so that the in-material keywords "PET" and "PA" can be selected together.

From the viewpoint similar to the above, each step in the mining process to be described later does not apply all the STEPs according to the system design method, but only some STEPs are applied or a combined / intersected relationship to be described later. It will be readily understood by one of ordinary skill in the art reading the present disclosure that the modifications may be different from those illustrated in FIGS. 6 to 8. Therefore, hereinafter, only the method (STEPs) illustrated in FIGS. 6 to 8 will be described for convenience and focus of the description, and a description thereof will be omitted for other various modifications.

According to the analysis method of the present invention, when the intellectual property documents are searched using the methodology described above, data mining and opinion mining for the search population are performed, and thus the correlation between the material keyword and the technical keyword is performed. Obtains material-technical association intellectual property documents that are evaluated as present (see methodology of FIG. 6), obtains material-emotional association intellectual property documents where association between material keyword and sentiment keyword is estimated to exist [Fig. Methodology of FIG. 7], thereby obtaining the material-technology-emotional-related intellectual property documents corresponding to the intersection of the material-technical-related intellectual property documents and the material-emotional-related intellectual property documents (FIG. 8). See methodology].

First, FIG. 6 illustrates an example of a data mining method for obtaining the material-technology related intellectual property documents. Each STEP is explained as follows.

[STEP 1] In the [Name of Invention] field of the data field of the intellectual property document, an intellectual property document including the material keyword and the technology keyword is obtained at the same time.

[STEP 2] In the [Summary] field of the data field of the IP document, an IP document including the material keyword is obtained.

[STEP 3] The material keyword and the technical keyword may be entered using at least one of the Summary, Technical Field, Effect of Invention, and Challenge fields in the data field of the IP document. Acquire an intellectual property document that is arranged adjacently within the first predetermined number of words in the front and rear.

Here, in FIG. 6, 'near' denotes the front and rear adjacent operators, and "near3" denotes the extraction of adjacent arrangements within three words back and forth (i.e., the material keyword and the description keyword are irrelevant to their arrangement order). Represents an expression.

The same applies to the following, but it may be appropriately selected by the system designer or the user to extract a contiguous arrangement within a few words. The larger the number of front and rear words or the back word as a reference condition for adjacent placement, the higher the reproducibility (ie, more relevant documents will be extracted), and the smaller the value, the higher the accuracy ( That is, only documents of higher relevance will be extracted), which is a matter of choice and irrelevant to the core technical principles of the present invention. Therefore, the numbers as reference conditions of the adjacent operation illustrated below make it clear that various modifications are possible.

[STEP 4] The description is based on the material keyword by using a rear neighbor operator in at least one of the [Summary], [Technical Field], [Effects of the Invention], and [Solution] fields of the data field of the intellectual property document. Obtain a intellectual property document in which a keyword is arranged adjacently within a second predetermined number of words behind.

Here, in FIG. 6, 'adj' denotes the backward adjoining operator, and “adj6” denotes an expression for extracting the adjacent arrangement of the description keyword within six words based on the material keyword.

[STEP 5] Intellectual Property Documents acquired in accordance with STEP 1, Intellectual Property Documents obtained in accordance with STEP 2 and acquired in accordance with STEP 2, and in accordance with STEP 4 simultaneously in accordance with STEP 2 Also, a first group of intellectual property documents obtained by unioning the obtained intellectual property documents is obtained.

[STEP 6] Key technical features of the intellectual property document by using forward and backward adjacent operators on at least one of the Summary, Technical Field, Effect of Invention, and Challenge fields of the data field. A pre-specified idiom keyword referring to the contents unique to the invention described as and the material keyword are acquired in the front and rear adjacently in a third predetermined number (four in this example) of the intellectual property document. For example, the idiom keyword may correspond to the present invention, the present invention, the core of the invention, and the like.

[STEP 7] Knowledge of simultaneously including the material keyword and the technology keyword in at least two of the [Summary], [Technical Field], [Effects of the Invention], and [Solution] fields in the data field of the intellectual property document. Obtain a property document.

[STEP 8] A second group of intellectual property documents acquired in accordance with STEP 2, and a second group of intellectual property documents acquired in accordance with STEP 6 and acquired in accordance with STEP 7, are acquired.

[STEP 9] The material-technology-related intellectual property documents are obtained by combining the first group acquired according to STEP 5 and the second group acquired according to STEP 8 in a union.

Next, FIG. 7 illustrates an example of an opinion mining method for obtaining the material-emotional associated intellectual property documents. Each STEP is explained as follows. Here, since [STEP 1] of FIG. 7 has already been described above as a synonym dictionary DB regarding emotion keywords, duplicate description thereof will be omitted.

[STEP 2] Knowledge of simultaneously including the material keyword and the emotional keyword in at least two of the [Summary], [Technical Field], [Effects of the Invention], and [Solution] fields in the data field of the intellectual property document. Obtain a property document.

[STEP 3] The material keyword and the emotional keyword are generated by using front and rear adjacent operators in at least one of the fields of [Summary], [Technical Field], [Effects of the Invention] and [Solution] among the data fields of the intellectual property document. Obtains the intellectual property document arranged adjacent to each other in the fourth predetermined number (five in this example) in the front and rear directions.

In addition, the emotional keyword is based on the material keyword by using a rear neighbor operator in at least one of the Summary, Technical Field, Effect of Invention, and Challenge fields of the data field of the intellectual property document. Acquiring the intellectual property document arranged adjacently within the fifth predetermined number of words (eight in this example) in the rear.

Thereafter, the intellectual property documents obtained according to the STEP 2, the intellectual property documents obtained according to the above two methods of STEP 3 are combined to obtain the material-emotional related intellectual property documents.

Accordingly, as shown in FIG. 8, the material-technology-emotional-related intellectual property documents which are the intersection of the material-technology-related intellectual property documents and the material-emotional-related intellectual property documents obtained as described above may be obtained. .

Subsequently, a material-technology-emotion matrix DB is constructed in which the obtained material-technology-emotion-related intellectual property documents are structured in a matrix form according to a material keyword, a technology keyword, and an emotion keyword. At this time, the step of generating the material-technology-sensitive matrix DB may be used as follows.

In general, in the material-technology-emotional-related intellectual property document, a plurality of technology keywords related to the material keyword and the emotion keyword may exist. Patent technology is often duplicated in the technical effect within a single document, it is necessary to efficiently manage the DB.

Accordingly, the material-technology-emotion matrix DB according to an embodiment of the present invention is arranged such that the plurality of description keywords are distributed and arranged in one data row when the material keyword and the emotion keyword are sorted and applied as a data filter value. It can be designed to have a variable matrix structure. An example of this is shown in FIGS. 9 and 10.

Here, FIG. 9 is an example of a variable matrix structure in which technical effects are dispersed and expressed on a matrix when mining data of a material-technology-emotional relation is matrixed, and FIG. 10 is obtained by searching for a technical effect and emotional effect matrix. This is an example of a material-related valid patent.

11 to 15 illustrate a comparison result between a general patent search method and an advanced design material discovery (AMD) method according to the present invention.

First, according to FIG. 11, in the general patent search method, a patent including all materials / technologies / sensitivity keywords as an intersection (that is, an AND condition) is searched, whereas in the AMD analysis method of FIG. The patent is extracted through the search and mining method of FIG. 8. Accordingly, in the general patent retrieval method, a large number of noises (that is, invalid data) as in the left figure of FIG. 14 are generated. This is because a large number of patents corresponding to simple intersections can be searched in addition to a highly related technology in which material and technology, material and emotion are strongly bound. For example, it is simply that the material keyword is included in a patent document and a large number of patent documents that are not used as essential contents of the patent are searched.

On the other hand, according to the AMD analysis method of the present invention, since it is highly likely to search for a highly related patent document strongly bound between material and technology, material and emotion, it can be confirmed that it is quite excellent in terms of accuracy (see the right figure in FIG. 14). ].

The performance of searching and mining by the AMD analysis method of the present invention can be confirmed through the calculation of the accuracy, recall, and a harmonic mean value related thereto described in FIG. 12. In this regard, the performance of the AMD analysis method of the present invention may be further improved by performing a sampling test on the above-mentioned performance evaluation indicators such as accuracy, recall, and harmonic mean value in advance. That is, as illustrated in FIG. 13, after combining the general search method and the patent documents acquired according to the AMD method of the present invention, the sampling test of the method of FIG. 15 is performed (ie, three performances calculated). By comparing the evaluation indicators, you can choose a more sensitive AMD method. For example, after performing a sampling test several times before and after the search and mining within a few words before and after as the operation condition according to the adjacent operator described above, the performance is better. That's what we use to select the actual computational conditions, given the performance metrics.

In addition to the above, the above description focuses on a case in which a search and mining using a single material keyword, a single technology keyword, and a single emotional keyword is performed. However, a plurality of keywords may be used for searching by attributes (material / technology / sensitivity). Of course you can. In this case, a plurality of keywords having the same attribute may be searched and mined in an intersection relationship, or may be searched and mined in a union relationship. That is, this may be different depending on how the system design is implemented.

Also, in some cases, after a designer or researcher completes the first search and mining using a specific material, technology, or emotion keyword, he or she may want to reapply additional keywords regarding at least one of the specific material, technology, and emotion keyword. In this case, the search and mining methods similar to those described above with reference to FIGS. 6 to 9 may be applied again. In this case, if additional search and mining is required in the material-technical-sensitivity-related intellectual property documents obtained primarily, re-searching and mining will be performed only on the first acquired intellectual property documents.

The big data analysis method based on the intellectual property document for design material search according to the embodiment of the present invention can be implemented as computer readable code on a computer readable recording medium. Computer-readable recording media include all kinds of recording media having data stored thereon that can be decrypted by a computer system. For example, there may be a read only memory (ROM), a random access memory (RAM), a magnetic tape, a magnetic disk, a flash memory, an optical data storage device, and the like. The computer readable recording medium can also be distributed over computer systems connected over a computer network, stored and executed as readable code in a distributed fashion.

Although the above has been described with reference to embodiments of the present invention, those skilled in the art may variously modify the present invention without departing from the spirit and scope of the present invention as set forth in the claims below. And can be changed easily.

Claims (7)

A computer implemented method for searching and analyzing design material based on intellectual property documents performed on a computer.
As a material keyword, a technology keyword, and an emotion keyword are received, performing a search in the intellectual property document DB using the received keyword—wherein the material keyword is a keyword indicating a name of the material, The technical keyword is a keyword expressing a functional feature determined according to the material property of the material, and the emotional keyword is a keyword expressing an external feature expressed according to the material property of the material; Obtaining material-technology-related intellectual property documents, which are evaluated to have an association between a material keyword and a technology keyword according to a predetermined criterion, on documents in the search population collected through the search; Acquiring material-emotional-related intellectual property documents, which are estimated to have an association between a material keyword and an emotional keyword according to a predetermined criterion, for documents in the search population collected through the search; Obtaining a material-technology-emotional associated intellectual property document corresponding to an intersection of the material-technical intellectual property documents and the material-emotional associated intellectual property documents; Generating a material-technology-emotion matrix DB structured in the form of a matrix according to the material keyword, technology keyword, and emotion keyword.
Acquiring the material-technology related intellectual property documents,
Applying a first method of acquiring an intellectual property document including the material keyword and the technical keyword in the [name of invention] field of the data field of the intellectual property document; Applying a second method of acquiring an intellectual property document including the location keyword in the [Summary] field of the data field of the intellectual property document; The material keyword and the technical keyword are pre-assigned in front and rear using at least one of a summary, a technical field, an effect of the invention, and a task in the data fields of the IP document. Applying a third manner of obtaining an intellectual property document disposed adjacent within a first number of words; The technical keyword is rearwarded based on the material keyword by using a rear adjacency operator in at least one of the [Summary], [Technical Field], [Effects of the Invention] and [Solution] fields of the data field of the IP document. Applying a fourth manner of obtaining an intellectual property document disposed adjacent in a second predetermined number of words; Inherent to the invention as described as a key technical feature in the intellectual property document using at least one of the Summary, Technical Field, Effect of Invention, and Challenge fields in the data field of the intellectual property document. Applying a fifth method of acquiring a pre-designated idiom keyword indicating the content of the subject matter and the subject keyword in the vicinity of the pre-designated third number of words; Acquiring an intellectual property document including the material keyword and the technical keyword simultaneously in at least two of the [Summary], [Technical Field], [Effects of the Invention], and [Solution] fields of the data field of the intellectual property document. Applying a sixth scheme to do so; IPR documents acquired in accordance with the first scheme, IPR documents acquired in accordance with the second scheme and acquired in accordance with the third scheme, acquired in accordance with the second scheme, and acquired in the fourth scheme. Acquiring the first group of intellectual property documents, the union of the obtained intellectual property documents; Acquiring a second group of intellectual property documents obtained according to the second method, and intersecting the intellectual property documents acquired according to the fifth method and also obtained according to the sixth method; And combining the first group and the second group of intellectual property documents to obtain the material-technical associated intellectual property documents,
Acquiring the material-emotional IP document,
Acquiring an intellectual property document that includes the material keyword and the emotional keyword simultaneously in at least two of the [Summary], [Technical Field], [Effects of the Invention], and [Solution] fields of the data field of the intellectual property document. Applying a seventh manner to do; The material keyword and the emotional keyword are pre-assigned in front and rear using at least one of a summary, technical field, effect of invention, and task in the data field of the IP document. Applying an eighth manner of obtaining an intellectual property document disposed adjacent within a fourth number of words; The emotional keyword is rearward based on the material keyword by using a rear neighbor operator in at least one of the Summary, Technical Field, Effect of Invention, and Challenge fields of the data field of the intellectual property document. Applying a ninth manner of acquiring the intellectual property document disposed adjacent in the fifth predetermined number of words; The material-emotional related intellectual property documents which combine the intellectual property documents acquired according to the seventh method, the intellectual property documents acquired according to the eighth method, and the intellectual property documents acquired according to the ninth method. And acquiring.
The method of claim 1,
A synonym dictionary DB (database), which is managed by grouping the synonyms associated with the keyword by the material keyword, the technical keyword, and the emotional keyword, is represented by a specific keyword.
And performing an extended search including the synonyms of the corresponding keyword according to the synonym dictionary DB in the process of searching for the intellectual property document using the keyword.
The method of claim 2,
When any one material keyword in the synonym dictionary DB regarding the material keyword includes a material keyword of a plurality of sub-concepts belonging to the keyword as a higher-level keyword, the keyword of the higher-concept and the lower-concept in the keyword dictionary DB Keywords are linked and managed,
And searching for the intellectual property document using the keyword, and performing an extended search including a material keyword of a subordinate concept managed in association with the corresponding material keyword.
delete delete delete The method of claim 1,
The step of generating the material-technology-sensibility matrix DB,
In the material-technological-related intellectual property document, when there are a plurality of technical keywords related to the material keyword and the emotional keyword,
The material-technology-emotion matrix DB has a variable matrix structure in which the plurality of technology keywords are distributed and arranged in one data row when the material keyword and the emotion keyword are sorted and applied as data filter values. , Computer implemented method.

Images