WO2011136413A1

WO2011136413A1 - Apparatus and method for configuring a comprehensive intellectual property rights star network by detecting patent similarity

Info

Publication number: WO2011136413A1
Application number: PCT/KR2010/002730
Authority: WO
Inventors: 강종석; 이혁재; 문영호
Original assignee: 한국과학기술정보연구원
Priority date: 2010-04-30
Filing date: 2010-04-30
Publication date: 2011-11-03

Abstract

The present invention relates to an apparatus and method for configuring a comprehensive intellectual property (IP) rights star network by detecting patent similarity, wherein the apparatus comprises: a patent data search and storage unit for accessing a patent database, searching for patent data corresponding to a search query, and downloading and storing data found from the search query; a patent data standardizing unit for standardizing the stored patent data according to respective analysis categories and generating a metadatabase by analysis category; a similarity computing unit whereby a simultaneous occurrence matrix by analysis category is derived by using the number of metadata that simultaneously occur among the respective patent data in the information by analysis category from the metadatabase, or by using a frequency of occurrence and deriving a similarity matrix wherein the similarities among the patent data by category are shown by using the simultaneous occurrence matrix; a comprehensive star network visualizing unit for selecting two categories from among the analysis categories and establishing an x-axis and a y-axis, selecting one of the patent data and disposing it at the point of origin, and then using the similarity matrix corresponding to the selected analysis category to visualize the positions, distances, or interconnections of the respective patent data on a comprehensive star network according to the similarities between the patent data disposed at the point of origin and the remaining patent data; and a control unit for controlling the visualization of similarities among the patent data through the comprehensive star network by accessing the patent database and finding and downloading patent data, generating a metadatabase by analysis category for the respective patent data, and using the metadatabase to extract a simultaneous occurrence matrix and a similarity matrix for each analysis category.

Description

Apparatus and method for implementing intellectual property comprehensive forming network by detecting patent similarity

The present invention relates to an apparatus and a method for implementing a intellectual property comprehensive forming network by detecting a patent similarity for objectively estimating the correlation between the similarity of the patent data and visualizing the integrated forming network.

As the industrial base shifts to information and knowledge, interest in and acquisition of Intellectual Property Rights such as patent rights, utility model rights, design rights, trademark rights, and copyrights is increasing.

Many domestic and foreign companies have a patent management organization and patent strategy to manage their intellectual property, and through successful patent strategies, companies can not only secure an exclusive position in the market, strengthen their competitiveness, increase their corporate value, Licensing secures new revenue streams.

Accordingly, maximization of income based on intellectual property, conversion to profit-generating business using patent portfolio, and achievement of business purpose are emerging as major issues in corporate management.

In particular, as disputes continue to occur based on intellectual property rights, including patents, an active patent acquisition strategy at the national level and a countermeasure against infringement claims are required. Similarity assessment between registered rights has taken an important place.

Patent rights, which constitute the largest proportion of intellectual property rights, are for legal protection of inventions, which are human mental creations, and the technical idea to be protected is described in the specification to apply for protection of rights through the patent application procedure and to be protected through examination process. Means the property rights registered with the right to

The patent-pending specification summarizes the subject matter of the invention, describes the name of the invention, and describes the prior art and the objects, configurations, and effects of the present invention in detail so that those skilled in the art can easily understand the scope of the claims. This is a written description that clearly and concisely describes the technical content to be protected.

In general, for the prior art research of patent rights, a method of searching a technology in a related field by combining keywords was used. However, the above prior art research has a problem in that the results differ depending on the searcher and the objectivity of the analysis results cannot be guaranteed.

According to the present invention, the patent data is standardized according to the analysis field to calculate the similarity, at least one of the analysis areas is set as a variable, and the objective of calculating the similarity between each patent data is visualized by a comprehensive molding network. An apparatus for implementing a property rights covering network, a method thereof, and a recording medium readable by an electronic device recording the same.

According to an aspect of the present invention, a patent data search and storage unit for accessing a patent database to search for, download, and store patent data corresponding to a search expression, and standardize the stored patent data according to each analysis field to analyze a meta database for each analysis field. Patent data standardization unit to generate the data, and from the information in the meta database for each analysis field to calculate the co-occurrence matrix for each analysis field using the number or frequency of meta data simultaneously generated between each patent data and by the analysis field using the co-generation matrix The similarity calculation part that calculates the similarity matrix representing the similarity between patent data, selects two fields among the analysis fields and sets them on the x-axis and y-axis, and selects any one of the patent data at the origin and then selects the selected analysis. Using the similarity matrix corresponding to the field Based on the similarity between the patent data placed at the point and the rest of the patent data, the comprehensive molding network visualization unit visualizes the position, distance or connection of each patent data with the comprehensive molding network, and accesses the patent database to search for and download the patent data. Patent similarity including a control unit that generates a meta database for each patent data for each patent field and calculates a co-occurrence matrix and a similarity matrix for each analysis field using the meta database to visualize the similarity between patent data in a comprehensive forming network. Provided is an intellectual property comprehensive forming network realization apparatus by detection.

The controller may periodically access the patent database to search and update the patent data corresponding to the search expression to calculate a similarity matrix between the patent data and to visualize the similarity between the patent data in a comprehensive forming network.

The patent data standardization unit may standardize the stored patent data into metadata that is a minimum unit including a word or a phrase according to an analysis field.

The similarity calculating unit may indicate the similarity between patent data using any one of the similarity coefficients including jacquard, inclusion, proxy meter, and cosine.

The apparatus for implementing an intellectual property encompassing network by detecting patent similarity further includes an input unit for inputting a search expression, and accesses a patent database through a wired / wireless communication network to search for patent data.

The field of analysis may include a keyword, an applicant, an applicant nationality, an inventor, an inventor nationality, a technical classification code, a word or a phrase included in a claim.

According to another aspect of the present invention, a patent data retrieval and storage unit for accessing a patent database and retrieving and downloading citations or cited patent data of interest patent data and citation patent data, and storing stored patent data in respective analysis fields. Patent data standardization unit to generate a meta database for each analysis field by standardization according to the standard, and by using the number or frequency of occurrence of meta data simultaneously generated between citation or cited patent data of interest and patent data of interest from the meta database information for each analysis field. Similarity calculation part that calculates the similarity matrix for each analysis field and calculates the similarity matrix for each analysis field that shows the similarity between citation or cited patent data of interest patent data and interest patent data using the simultaneous generation matrix for each analysis field. Selecting one or more similarity matrices among the matrices and using the selected similarity matrix, the position, distance, or connection of each patent data into the comprehensive forming network according to the similarity between citations of the patent data of interest and patent data cited or cited. A comprehensive forming network visualizing unit that visualizes and accesses a patent database, retrieves and downloads patent data of interest and citation or cited patent data of interest patent data, creates a meta database for each analysis field, and uses a meta database for each patent data. Intellectual Property Rights by Patent Similarity Detection including a control unit that calculates the co-occurrence matrix and the similarity matrix for each analysis field to visualize the similarity between the patent data of interest and the citation or cited patent data of interest by a comprehensive forming network.The parenthesis star network implementation is provided.

The comprehensive forming network visualization unit selects one of the similarity matrix for each analysis field and cites or avoids the patent data of interest and the patent data of interest according to the similarity between the patent data of interest and the patent data of interest cited in the selected similarity matrix. The position, distance, or connection of the cited patent data can be visualized.

The comprehensive forming network visualization unit selects two of the analysis fields, sets the x-axis and the y-axis, respectively, and places the patent data of interest at the origin, and then the patent data of interest and the patent data of interest appearing in the similarity matrix corresponding to the selected analysis field. According to the similarity between the cited or cited patent data of the citation or cited patent data of interest patent data and the cited patent data can be visualized on the coordinate plane.

The controller periodically accesses the patent database, searches for and updates patent data corresponding to the search expression, calculates a similarity matrix between the patent data of interest and the citation or cited patent data, and visualizes the similarity between the patent data in a comprehensive forming network. Can be controlled.

The apparatus for implementing an intellectual property comprehensive forming network by detecting similarity of patents further includes an input unit for inputting patent information of interest, and accessing a patent database through wired / wireless communication networks to cite or cite patent data of interest and patent data of interest. You can search for data.

According to another aspect of the present invention, accessing a patent database, inputting a search expression, searching for patent data corresponding to the search expression from the patent database, downloading and integrating the searched patent data, converting the data into a set format, and storing the same. Generating a meta database for each analysis field by standardizing the stored patent data according to each analysis field, and using the number or frequency of occurrence of metadata simultaneously occurring between each patent data from the information of the meta database for each analysis field. Calculating a co-occurrence matrix for each analysis field, calculating a similarity matrix for each analysis field that shows similarity between patent data using the co-occurrence matrix for each analysis field, and selecting a similarity matrix for each analysis field for the x-axis or y-axis. Set it to a variable A method for implementing an intellectual property comprehensive forming network by detecting similarity between patents, comprising: placing one of the dummy data at the origin and visualizing the similarity between the patent data disposed at the origin and the remaining patent data with the comprehensive forming network on the coordinate plane To provide.

According to another aspect of the present invention, accessing a patent database to input interest patent data information, searching for citation or cited patent data of interest patent data and interest patent data corresponding to the interest patent data information from the patent database; Downloading, integrating, citation or cited patent data of interest patent data, patent data of interest, and converting and storing them into a set format, and standardizing the stored patent data according to each analysis field to generate a meta database for each analysis field. Calculating a co-occurrence matrix for each field of analysis using the number or frequency of occurrence of meta data simultaneously between the patent data of interest and citation or cited patent data of interest patent data from the information of the meta-data for each analysis field, analysis field Calculating a similarity matrix for each field of analysis, which indicates the similarity between the patent data of interest and the cited or cited patent data of the patent data of interest, and selecting one or more similarity matrices among the similarity matrix for each field of analysis and selecting the selected similarity matrix. Indicating by patent similarity detection comprising visualizing the position, distance or connection relationship of each patent data into a comprehensive forming network according to the similarity between the patent data of interest and the citation or cited patent data of the patent data of interest using Provides a method for implementing a comprehensive property network.

Selecting one or more similarity matrices among similarity matrices by analysis field and using the selected similarity matrices to cover the position, distance, or linkage of each patent data according to the similarity between citation of patent data of interest and patent data of citation or cited patent data In the step of visualizing the forming network, selecting two similarity matrices for each analysis field, setting the corresponding analysis field as a variable on the x-axis or y-axis, placing the patent data of interest at the origin, and then The similarity between the cited or cited patent data can be visualized by a comprehensive forming network on the coordinate plane.

In the step of accessing the patent database and inputting a search expression, the search expression may be generated by combining a keyword, a technology classification code, a patent registration number or a patent application number of the patent data to be searched.

Standardizing the stored patent data according to each analysis field to generate a meta-database for each analysis field includes: segmenting the stored patent data into words having a minimum unit according to each analysis field, and processing natural language processing of the stored patent data. Creating a list of words and phrases that have been segmented and naturalized, applying a thesaurus to the list, selecting meaningful metadata from words and phrases by thesaurus, and grouping them into groups of similar meaning And storing the metadata in a grouped thesaurus region as a meta database.

Computing a co-occurrence matrix using the number or frequency of occurrence of meta data simultaneously from the information of the meta-database for each analysis field may include extracting a minimum semantic unit including words and phrases from the meta database, and a minimum semantic unit. It can be configured to include the step of calculating the frequency of occurrence of.

The method of implementing an IPR-type forming network by detecting similarity of patents periodically accesses a patent database, searches for and updates citation or cited patent data corresponding to the patent data information of interest, and calculates a similarity matrix between patent data and between patent data. The method may further include controlling the similarity of to be visualized by the comprehensive forming network.

According to another aspect of the present invention, a program for performing an IPR implementation method by detecting patent similarity may be recorded and recorded on a recording medium readable by an electronic device.

According to the present invention, the similarity between patent data can be detected by objective criteria, and the similarity and degree of similarity between a plurality of patent data can be grasped at a glance.

In addition, the similarity of citation or cited patent data related to specific patent data can be calculated, so that the possibility of infringement of rights can be objectively analyzed and the dispute can be reconciled in advance.

In addition, the blank technology can be derived from the visible comprehensive forming network, and it can be used as a reference for setting research directions in related fields.

1 is a block diagram showing an embodiment of the intellectual property comprehensive forming network implementation apparatus by the patent similarity detection according to an aspect of the present invention.

Figure 2 is a connection relationship between the intellectual property comprehensive forming network implementation apparatus and the patent database by detecting the patent similarity according to an aspect of the present invention.

Figure 3 is a flow diagram illustrating an embodiment of a method for implementing an IPR forming network by detecting patent similarity according to another aspect of the present invention.

Figure 4 is a flow diagram illustrating an embodiment of a method for generating a meta database of a method for implementing an IPR forming network by detecting patent similarity according to another aspect of the present invention.

5 is a flowchart illustrating an embodiment of a method for calculating a co-occurrence matrix of a method for implementing an intellectual property encompassing forming network by detecting patent similarity according to another aspect of the present invention.

Figure 6 is a comprehensive forming network visualized by one embodiment of a method for implementing an intellectual property comprehensive forming network by the patent similarity detection of another aspect of the present invention.

Figure 7 is a comprehensive forming network visualized by another embodiment of a method for implementing an intellectual property comprehensive forming network by the patent similarity detection of another aspect of the present invention.

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 modifications, equivalents, and substitutes included in the spirit and scope of the present invention.

Now, an apparatus for implementing an intellectual property comprehensive forming network by detecting patent similarity according to an embodiment of the present invention, a method thereof, and a recording medium readable by an electronic device recording the same will be described in detail with reference to the drawings. The same or corresponding elements will be given the same reference numerals and redundant description thereof will be omitted.

<특허 유사도 검출에 의한 지적재산권 포괄 성형망 구현장치><Intellectual Property Comprehensive Forming Network Implementation Device by Patent Similarity Detection>

1 is a block diagram showing an embodiment of an intellectual property comprehensive forming network implementation apparatus 100 by detecting a patent similarity according to an aspect of the present invention, Figure 2 is a patent similarity detection according to an aspect of the present invention FIG. 3 is a diagram illustrating a connection relationship between the IPR apparatus 100 and the patent database 300.

According to the exemplary embodiment of the present invention, as shown in FIG. 1, the apparatus 100 for implementing an intellectual property comprehensive forming network by detecting patent similarity may include a patent data search and storage unit 110, a patent data standardization unit 120, It includes a similarity calculation unit 130, a comprehensive forming network visualization unit 140, the controller 150 and the input unit 160.

In addition, as shown in FIG. 2, the apparatus 100 for implementing the IPR by detecting patent similarity may be connected to the patent database 300 through a wired / wireless communication network 200 such as the Internet.

Hereinafter, the features of each component will be described in detail through various embodiments.

First embodiment

According to the first embodiment of the present invention, the patent data retrieval and storage unit 110 may access the patent database 300, search for and integrate the patent data corresponding to the search expression, download it, convert it into a set format, and store it. .

The search expression may be composed of keywords, applicants, inventors, technical classification code, application number, publication number, registration number or a combination of the above, the patent database 300 is a database established by the Patent Office of each country or commercially available patent database It can be configured to include.

The patent data standardization unit 120 may generate a meta database for each analysis field by standardizing the stored patent data into metadata that is a minimum unit including a word or a phrase according to each analysis field.

The field of analysis may include a keyword, an applicant, an applicant nationality, an inventor, an inventor nationality, a technical classification code, a word or phrase included in a claim, and may be applied as a criterion for determining similarity of patent data.

In general, a patent application specification is composed of a combination of words and phrases including letters and special symbols, the patent application is in addition to the above patent application specification bibliographic information for applicants, inventors, etc. and technical classification symbols according to the technical field The similarity of the scope of patent rights can be extracted by comparing and analyzing the matters described in the above-described patent application.

The similarity calculation unit 130 analyzes the number or frequency of occurrence of meta data simultaneously between each patent data from the information of the meta database for each analysis field, and collects the above analysis results to calculate a co-occurrence matrix for each analysis field. The similarity matrix can be calculated using the field-specific co-occurrence matrix.

Specifically, as shown in [Table 1], the simultaneous matrix for each analysis field is arranged by arranging a plurality of lists of patent data in a row or a column, and by placing metadata of the analysis field in a column or a row, each metadata is included in the patent data. It means a matrix indicating the number or frequency of occurrence or as shown in [Table 2], a plurality of patent data list arranged in a row and a column, respectively, and a matrix indicating the number or frequency of the corresponding meter data.

Table 1

	D1	D2	D3	D4	D5	D6	D7	D8	D9	D10
Tech1	One	0	One	One	One	One	0	One	One	0
Tech2	One	One	One	One	0	0	One	One	One	One
Tech3	0	0	One	One	One	One	One	0	One	One
Tech4	One	0	0	One	0	One	One	One	One	0
Tech5	0	One	0	One	One	One	0	One	One	One

TABLE 2

	D1	D2	D3	D4	D5	D6	D7	D8	D9	D10
D1	<3>
D2	One	<2>
D3	2	One	<3>
D4	3	2	3	<5>
D5	One	One	2	3	<3>
D6	2	One	2	4	3	<4>
D7	2	One	2	3	One	2	<3>
D8	3	2	2	4	2	3	2	<4>
D9	3	2	3	5	3	4	3	4	<5>
D10	One	2	2	3	2	2	2	2	3	<3>

From the co-occurrence matrix, we can estimate the correlation and comprehensiveness of patent data on related technologies.

For example, as shown in [Table 1], D1 patent data and D2 patent data are only related to Tech2 technology, and D1 patent data and D3 patent data are estimated to be related to Tech1 and Tech2 technology. The description of D1 patent data can be assumed to be included in the description of D4 patent data.

In addition, the similarity calculation unit 130 may calculate a similarity matrix for each analysis field indicating the similarity between the respective patent data using the simultaneous generation matrix for each analysis field.

The similarity constituting the similarity matrix may be calculated using any one of similarity coefficients including jacquard, inclusion, proxy meter, cosine, and the like.

[Table 3] shows and calculated the similarity matrix from the co-occurrence matrix of [Table 2], the similarity between the D1 to D10 patent data is calculated using the Jacquard similarity coefficient.

TABLE 3

	D1	D2	D3	D4	D5	D6	D7	D8	D9	D10
D1	<1>
D2	1/4	<1>
D3	2/3	1/4	<1>
D4	3/5	2/5	3/4	<1>
D5	1/5	1/4	1/2	3/5	<1>
D6	2/5	1/5	2/5	4/5	3/4	<1>
D7	1/2	1/4	1/2	3/5	1/5	2/5	<1>
D8	3/4	1/2	2/5	4/5	2/5	3/5	2/5	<1>
D9	3/5	2/5	3/5	One	3/5	4/5	3/5	4/5	<1>
D10	1/5	2/3	2/3	3/5	1/2	2/5	2/3	2/5	3/5	<1>

Jacquard similarity coefficient may be calculated by Equation 1.

Equation 1

In addition, the similarity matrix may be calculated using the inclusion similarity coefficient of [Equation 2] and the proxy meter similarity coefficient of [Equation 3].

Equation 2

Equation 3

The comprehensive forming network visualization unit 140 selects two similarity matrices among the similarity matrices for each analysis field, sets the analysis field corresponding to the selected similarity matrix as the x-axis and the y-axis, respectively, and selects any one of the patent data and places it at the origin. In addition, the remaining patent data may be disposed on the coordinate plane according to the similarity between the patent data disposed at the origin and the remaining patent data to visualize the comprehensive forming network.

The comprehensive forming network visualization unit 140 may select two analysis fields from among analysis fields, set them as x and y variables, and select reference patent data among a plurality of patent data and place them at the origin. In this case, coordinates of the remaining patent data are determined according to the similarity between the reference patent data disposed at the origin and the remaining patent data for the two selected analysis fields.

The controller 150 controls the patent data retrieval and storage unit 110, the patent data standardization unit 120, the similarity calculator 130, and the comprehensive forming network visualization unit 140.

Specifically, the controller 150 accesses the patent database 300, searches for and downloads patent data, creates a meta database for each analysis field for each patent data, and uses the meta database to analyze the co-occurrence matrix and similarity for each analysis field. By calculating the matrix, the similarity between the patent data can be controlled to be visualized by the comprehensive forming network.

In addition, the controller 150 periodically accesses the patent database 300 to search and update the patent data corresponding to the search expression to calculate a similarity matrix between the patent data and to visualize the similarity between the patent data in a comprehensive forming network. Can be controlled.

The apparatus 100 for implementing an intellectual property encompassing network by detecting patent similarity may further include an input unit 160 through which a search expression is input.

Second embodiment

According to the second embodiment of the present invention, the patent data retrieval and storage unit 110 may access the patent database 300 to retrieve and download and store citations or citations of patent data of interest and patent data of interest. .

The patent data retrieval and storage unit 110 of the first embodiment retrieves and downloads the patent data corresponding to the search expression uniformly, while the patent data retrieval and storage unit 110 of the second embodiment retrieves the specified interest patent data and interest. By retrieving and downloading the citation or cited patent data of the patent data, the second embodiment is for calculating the relation between the patent data of interest and the citation or citation patent data with high similarity.

The patent data standardization unit 120, the similarity calculation unit 130, the comprehensive forming network visualization unit 140, and the controller 150 are similar to the functions described in the first embodiment, and the similarity calculation unit 130 is the patent data of interest. And the difference in calculating the co-occurrence matrix or the similarity matrix between the cited or cited patent data of the patent data of interest, and the comprehensive forming network visualization unit 140 is between the cited or cited patent data of the patent data of interest and the patent data of interest. According to the similarity of, there is a difference in visualizing the position, distance or connection relationship of each patent data by the comprehensive forming network.

When the comprehensive forming network visualization unit 140 selects one of the similarity matrix for each analysis field, the patent data of interest and the patent data of interest according to the similarity between the patent data of interest and the citation or cited patent data of the patent data of the selected similarity matrix are selected. The location, distance or connection of the citation or citation of the cited patent data can be visualized.

That is, the degree of similarity between citations or citations of patent data of interest and patent data of interest can be grasped in one dimension.

In addition, when the comprehensive forming network visualization unit 140 selects two of the analysis fields, each field is set to the x-axis and the y-axis, and the patent data of interest is disposed at the origin, and then the similarity matrix corresponding to the selected analysis field is selected. According to the similarity between the patent data of interest and the citation or cited patent data of the patent data of interest, the position, distance or connection of the citation or citation of the patent data of interest and the cited patent data can be visualized on the coordinate plane. .

In this case, the similarity between the citations of the patent data of interest and the patent data of interest or the cited patent data can be grasped in two dimensions.

<특허 유사도 검출에 의한 지적재산권 포괄 성형망 구현방법><Method of Implementing Intellectual Property-Inclusive Molding Network by Patent Similarity Detection>

3 is a flowchart illustrating an embodiment of a method for implementing an IPR forming network by detecting patent similarity according to another aspect of the present invention, and FIG. 4 is an IPR forming network by detecting patent similarity according to another aspect of the present invention. FIG. 5 is a flowchart illustrating an embodiment of a method for generating a meta database of an implementation method, and FIG. 5 is a flowchart illustrating an embodiment of a method of calculating a concurrent matrix of a method for implementing an IPR-type comprehensive network by detecting patent similarity according to another aspect of the present invention. to be.

According to the present embodiment, as shown in FIG. 3, the method for implementing an intellectual property comprehensive forming network by detecting patent similarity may be performed by inputting a search formula (S310), searching for patent data (S320), downloading and integrating the same. The format is converted and stored (S330), a meta-database is generated (S340), a simultaneous matrix is calculated (S350), a similarity matrix is calculated (S360), and visualized by a comprehensive forming network (S370).

First embodiment

According to the first embodiment of the present invention, when a search expression is input from the input unit by accessing the patent database (S310), the patent data corresponding to the search expression is searched (S320), and the searched patent data is downloaded and integrated into a set format. Convert and store (S330).

The search expression may be generated by combining keywords or technical classification codes of patent data to be searched, and may be generated by combining bibliography of a publication or a registered publication.

Next, the stored patent data may be standardized according to each analysis field to generate a meta database for each analysis field (S340).

Meta database generation (S340) is to segment the stored patent data into the word of the minimum unit according to each analysis field (S332), after processing the stored patent data with natural language (S333), and the words and phrases processed with segmentation and natural language By creating a list of (S334), the thesaurus method is applied (S335).

The meta database may be generated by selecting meaningful meta data among words and phrases by the thesaurus, grouping them into groups of similar meaning (S336), and storing the meta data in the thesaurus region as a meta database (S340).

After generating the meta database (S340), the simultaneous generation matrix for each analysis field is calculated using the number or frequency of occurrence of the meta data simultaneously occurring between the patent data from the information of the meta database for each analysis field (S350).

Simultaneous matrix calculation may be represented by searching the meta database (S342), extracting the minimum meaning unit including words and phrases from the meta database (S344), and calculating the frequency of occurrence of the minimum meaning unit (S350).

A similarity matrix indicating the similarity between patent data is calculated using the simultaneous generation matrix for each analysis field (S360), and the similarity between each patent data can be visualized by the comprehensive forming network (S370).

In the process of visualizing the comprehensive forming network (S370), when one of the similarity matrices for each analysis field is selected, each of the selected patent data may be adjusted according to the similarity indicated in the similarity matrix.

In the process of visualizing with a comprehensive forming network (S370), when two of the similarity matrices for each analysis field are selected, each analysis field is set as an x and y variable, and one of the patent data is placed at the origin and then placed at the origin. The position, distance or connection between the patent data and the rest of the patent data can be adjusted to visualize a comprehensive forming network on the coordinate plane.

In addition, when the set period elapses (S380), the patent database is accessed to search for and update the citation or cited patent data corresponding to the patent data information of interest to calculate a similarity matrix between the patent data, and to cover the similarity between the patent data. It can be visualized by the forming network.

Second embodiment

The second embodiment of the present invention visualizes the similarity between the citations of the patent data of interest and the patent data of interest or the cited patent data by a comprehensive forming network through a similar process to each step of the first embodiment, and thus the description will be mainly focused on the differences. .

According to the second embodiment of the present invention, when the interest patent data information is input from the input unit by accessing the patent database (S310), the citation or citation of the patent data of interest and the patent data of interest is searched (S320), and the retrieved patent data. The patent data is downloaded and integrated into a set format and stored (S330).

In the first embodiment, a plurality of patent data corresponding to the search expression is searched uniformly, while in the second embodiment, the patent data of interest is primarily searched, and the citation or cited patent data of the patent data of interest is secondarily searched. There is a difference that is retrieved.

Simultaneous matrix or similarity matrix calculation calculates the patent data of interest and patent data of citation or cited patent data for each analysis field, and the comprehensive molding network visualization selects one or more similarity matrix among similarity matrix by analysis field and selects similarity. The matrix is used to visualize the position, distance, or connection relationship of each patent data into a comprehensive forming network according to the similarity between the patent data of interest and the citation or cited patent data of the patent data of interest.

That is, the first embodiment visualizes the similarity between a plurality of patent data as a comprehensive molding network, while the second embodiment uses the similarity between the citation or cited patent data of the patent data of interest and the patent data of interest as the comprehensive molding network. There is a difference in visibility.

In the process of visualizing the comprehensive forming network, if two of the similarity matrices are selected for each analysis field, the corresponding analysis field is set as the variable of the x-axis or the y-axis, and the patent data of interest is placed at the origin, and then the patent data of interest and interest The similarity between citation of patent data or cited patent data can be visualized two-dimensionally on the coordinate plane.

6 is a comprehensive forming network visualized by one embodiment of a method for implementing an intellectual property comprehensive forming network by patent similarity detection of another aspect of the present invention, and FIG. 7 is an intellectual property right by patent similarity detection of another aspect of the present invention. FIG. 3 is a view illustrating a comprehensive forming network visualized by another embodiment of a comprehensive forming network implementation method.

As shown in FIG. 6, the similarity between the patent data P of interest and the citations of the patent data of interest (B1 to B7) or the citations of the cited patents (F1 to F7) can be visualized in a comprehensive speech network in one analysis field. Can be.

The patent data is represented by a node, and the similarity between the patent data may be represented by a connection relationship, and when the similarity is 0.5 or more, it may be visualized to be connected by a line.

As shown in FIG. 7, the similarity between the patent data of interest and the citation or cited patent data of the patent data of interest can be visualized two-dimensionally in two fields of analysis.

Here, the x coordinate of the cited or cited patent data means the similarity with respect to the field of x analysis of the patent data of interest, and the y coordinate means the similarity with respect to the y analysis field of the patent data of interest.

Since the first cited or cited patent data is closer to the patent data of interest than the second cited or cited patent data, the similarity with the patent data of interest is high in the field of x and y analysis, and the first cited patent data is analyzed by x The similarity with the patent data of interest is higher in the y analysis field than the field, and the first cited patent data may be estimated to have a similarity with the patent data of interest in the x analysis field than the y analysis field.

In addition, the second citation patent data can be estimated to have the lowest similarity with the patent data of interest in the x, y analysis field.

<특허 유사도 검출에 의한 지적재산권 포괄 성형망 구현방법이 기록된 전자 장치에서 판독 가능한 기록매체><Recordable Media Readable by Electronic Device Recording Method for Implementing IP-Inclusive Forming Network by Patent Similarity Detection>

The intellectual property comprehensive forming network implementation method by detecting the patent similarity can be written by a computer program, and codes and code segments constituting the program can be easily inferred by a computer programmer in the art.

In addition, the program related to the implementation of the intellectual property comprehensive forming network by detecting the patent similarity is stored in a computer readable medium, and read and executed by the computer to capture the similarity of the patent data. Visualization can be done with

The present invention can be applied to an apparatus for implementing a comprehensive forming network of intellectual property rights by patent similarity detection.

In addition, the present invention can be applied to fields related to derivation of technology or prevention of patent infringement requiring research and development according to patent similarity detection.

Claims

A patent data search and storage unit for accessing a patent database to search for, download and store patent data corresponding to a search expression;

A patent data standardization unit generating a meta database for each analysis field by standardizing the stored patent data according to each analysis field;

From the information of the meta-database for each analysis field, a simultaneous matrix for each analysis field is calculated using the number or frequency of metadata occurring simultaneously between each patent data, and the patent data for each analysis field is calculated using the co-occurrence matrix. A similarity calculation unit for calculating a similarity matrix indicating a similarity degree;

Select two fields among the analysis fields and set them on the x-axis and y-axis, and select any one of the patent data and place them at the origin, and then place them at the origin by using the similarity matrix corresponding to the selected analysis field. A comprehensive forming network visualization unit for visualizing the position, distance, or connection relationship of each patent data into a comprehensive forming network according to the similarity between the patent data and the remaining patent data; And

Access to the patent database, search for and download patent data, create a meta database for each field of analysis for each patent data, and calculate the co-occurrence matrix and similarity matrix for each field of analysis using the meta database to obtain similarity between the patent data. Apparatus for implementing a intellectual property comprehensive forming network by detecting a patent similarity comprising a; control unit to control the visualization to a comprehensive forming network.
The method of claim 1,

The controller periodically accesses the patent database, searches for and updates patent data corresponding to the search expression, calculates a similarity matrix between the patent data, and controls to visualize the similarity between the patent data in a comprehensive forming network. Apparatus for implementing an intellectual property comprehensive forming network by detecting similarity.
The method of claim 1,

The patent data standardizing unit implements the intellectual property comprehensive forming network by patent similarity detection, characterized in that the standardized to the metadata stored in the minimum unit containing the word or phrase according to the analysis field.
The method of claim 1,

The similarity calculating unit implements the intellectual property comprehensive forming network by detecting a patent similarity indicating the similarity between the patent data using any one of the similarity coefficients including jacquard, inclusion, proxy meter, cosine.
The method of claim 1,

And an input unit for inputting the search expression, and accessing the patent database through a wired / wireless communication network to search for the patent data.
The method of claim 1,

The analysis field includes a keyword, applicant, applicant nationality, inventor, inventor nationality, technical classification code, a word or phrase included in the claims, the intellectual property comprehensive forming network implementation apparatus by the patent similarity detection.
A patent data retrieval and storage unit for retrieving, downloading, and storing patent data of interest and citation or cited patent data of the patent data of interest by accessing a patent database;

A patent data standardization unit generating a meta database for each analysis field by standardizing the stored patent data according to each analysis field;

From the meta database information for each analysis field, a simultaneous generation matrix for each analysis field is calculated by using the number or frequency of occurrence of metadata between the patent data of interest and the citation or cited patent data of the patent data of interest. A similarity calculation unit configured to calculate a similarity matrix for each analysis field indicating a similarity between the patent data of interest and citation or cited patent data of the patent data of interest using a generation matrix;

Selecting one or more similarity matrices of the similarity matrix for each analysis field and using the selected similarity matrix, the position, distance, or location of each patent data according to the similarity between the patent data of interest and the citation or cited patent data of the patent data of interest. Comprehensive forming network visualization unit for visualizing the connection relationship to the forming network; And

Access the patent database, search for and download the patent data of interest and citation or cited patent data of the patent data of interest, and generate a meta database for each analysis field for each patent data and simultaneously generate the analysis field by using the meta database. A control circuit for calculating a matrix and a similarity matrix to control the visualization of the similarity between the patent data of interest and the citation or cited patent data of the patent data of interest by a comprehensive shaping network; Implementation device.
The method of claim 7, wherein

The comprehensive forming network visualization unit may select one of the similarity matrix for each analysis field, and according to the similarity between the patent data of interest and the citation or cited patent data of the patent data of interest shown in the selected similarity matrix, An apparatus for implementing an intellectual property comprehensive forming network by detecting similarity of a patent that visualizes the position, distance, or connection relation of citation or cited patent data of interest.
The method of claim 7, wherein

The comprehensive forming network visualizing unit selects two of the analysis fields, sets the x-axis and the y-axis, respectively, and places the patent data of interest at an origin, and then displays the patent of interest shown in the similarity matrix corresponding to the selected analysis field. Patent similarity detection which visualizes the position, distance or connection relationship of the citation or cited patent data of the patent data of interest and the cited patent data on a coordinate plane according to the similarity between the data and the citation or cited patent data of the patent data of interest. An apparatus for implementing a comprehensive network of intellectual property rights by means of.
The method of claim 7, wherein

The controller periodically accesses the patent database, searches for and updates patent data corresponding to the search expression, calculates a similarity matrix between the patent data of interest and the citation or cited patent data, and calculates the similarity between the patent data. An apparatus for implementing an intellectual property comprehensive forming network by detecting a patent similarity which controls the visualization to a comprehensive forming network.
The method of claim 7, wherein

The similarity calculating unit implements the intellectual property comprehensive forming network by detecting a patent similarity indicating the similarity between the patent data using any one of the similarity coefficients including jacquard, inclusion, proxy meter, cosine.
The method of claim 7, wherein

The apparatus further includes an input unit through which the patent data information of interest is input, and by accessing the patent database through a wired / wireless communication network to detect the citation or cited patent data of the patent data of interest and patent data of interest. Device for implementing a comprehensive forming network of intellectual property rights.
The method of claim 7, wherein

The patent data standardizing unit implements the intellectual property comprehensive forming network by patent similarity detection, characterized in that the standardized to the metadata stored in the minimum unit containing the word or phrase according to the analysis field.
The method of claim 7, wherein

The analysis field includes a keyword, applicant, applicant nationality, inventor, inventor nationality, technical classification code, a word or phrase included in the claims, the intellectual property comprehensive forming network implementation apparatus by the patent similarity detection.
Accessing a patent database and inputting a search expression;

Retrieving patent data corresponding to the search expression from the patent database;

Downloading and integrating the retrieved patent data into a predetermined format and storing the converted patent data;

Generating a meta database for each analysis field by standardizing the stored patent data according to each analysis field;

Calculating a co-occurrence matrix for each analysis field by using the number or frequency of metadata occurring simultaneously between respective patent data from the information of the meta-database for each analysis field;

Calculating a similarity matrix for each analysis field indicating the similarity between the patent data using the analysis field-specific co-occurrence matrix; And

Select two of the similarity matrices for each analysis field as variables of x-axis or y-axis, and place any one of the patent data at the origin, and then measure the similarity between the patent data disposed at the origin and the remaining patent data on the coordinate plane. Visualizing a comprehensive forming network; Intellectual property comprehensive forming network implementation method by the patent similarity detection comprising a.
Accessing a patent database and inputting patent data information of interest;

Retrieving patent data of interest corresponding to the patent data of interest and citation or cited patent data of the patent data of interest from the patent database;

Downloading and integrating the retrieved patent data of interest, citations or citations of the patent data of interest, and converting and storing the patent data of interest;

Generating a meta database for each analysis field by standardizing the stored patent data according to each analysis field;

Calculating a co-occurrence matrix for each analysis field by using the number or frequency of occurrence of metadata simultaneously occurring between the patent data of interest and citation or cited patent data of the patent data of interest from the information of the analysis field-specific meta database;

Calculating a similarity matrix for each analysis field indicating a similarity between the patent data of interest and citation or cited patent data of the patent data of interest using the simultaneous generation matrix for each analysis field; And

Selecting one or more similarity matrices among the similarity matrices for each analysis field and using the selected similarity matrices, the position, distance, or location of each patent data according to the similarity between the patent data of interest and citation or cited patent data of the patent data of interest. Implementing a comprehensive network of intellectual property rights by detecting the patent similarity comprising the;
The method of claim 16,

Selecting one or more similarity matrices of the similarity matrix for each analysis field and using the selected similarity matrix, the position, distance, or location of each patent data according to the similarity between the patent data of interest and the citation or cited patent data of the patent data of interest. The step of visualizing the connection relationship as a comprehensive forming network,

After selecting two of the similarity matrix for each analysis field, the corresponding analysis field is set as a variable on the x-axis or y-axis, and the patent data of interest is placed at the origin, and then the citation or citation of the patent data of interest and the patent data of interest is performed. Implementation method of intellectual property comprehensive forming network by detecting patent similarity which visualizes similarity between patent data to comprehensive forming network on coordinate plane
The method according to claim 15 or 16,

In the step of accessing the patent database and entering a search expression,

The search formula is an intellectual property comprehensive forming network implementation method by detecting the patent similarity generated by combining a keyword, technology classification code, patent registration number or patent application number of the patent data to search.
The method according to claim 15 or 16,

Standardizing the stored patent data according to each analysis field to generate a meta database for each analysis field,

Dividing the stored patent data into words having a minimum unit according to each analysis field;

Natural language processing the stored patent data;

Creating a list of words and phrases processed with the segment and natural language;

Applying a thesaurus method to the created list;

Selecting meaningful metadata of words and phrases by the thesaurus and grouping them into groups of similar meaning; And

And storing the grouped thesaurus region as a meta database.
The method according to claim 15 or 16,

Calculating a co-occurrence matrix using the number or frequency of meta data occurring at the same time from the information of the meta-database for each analysis field,

Extracting a minimum semantic unit including words and phrases from the meta database; And

Comprising: calculating the frequency of occurrence of the minimum meaning unit; Intellectual property comprehensive forming network implementation method by the patent similarity detection comprising a.
The method according to claim 15 or 16,

Periodically access the patent database to search and update the citation or citation patent data corresponding to the patent data information of interest to calculate a similarity matrix between the patent data and control the visualization of the similarity between the patent data to a comprehensive forming network Implementing a comprehensive network encompassing intellectual property rights by the patent similarity detection further comprising.
22. The recording medium according to any one of claims 15 to 21, wherein a program for performing a method for implementing an intellectual property comprehensive forming network by detecting patent similarity is recorded and can be read by an electronic device.