A method for processing intellectual property data

Abstract

The application provides a kind of intellectual property data processing method, it is related to intellectual property service technical field, including the following steps: to intellectual property data is cleaned;Extract the target and keyword of each intellectual property manuscript;Association target, keyword and intellectual property manuscript, generate target-intellectual property atlas;When the target of intellectual property manuscript includes at least two application subjects, application subject is used as collaborative subject, and intellectual property manuscript is used as collaborative manuscript;From target-intellectual property atlas, the keyword of each intellectual property manuscript of collaborative subject is obtained, and the contribution area of each collaborative subject on collaborative manuscript is determined;The number of intellectual property manuscripts of collaborative subject is counted, and the first correlation degree between collaborative subjects is determined.The related elements of intellectual property are associated, and on this basis, the contribution area of each collaborative subject is determined, the correlation relationship and respective technical contribution of different subjects in collaborative research and development are intuitively presented, and the research and development correlation degree between collaborative subjects is quantified.

Description

Technical Field

[0001] This invention relates to the field of intellectual property service technology, and specifically to a method for processing intellectual property data. Background Technology

[0002] With the deepening development of the knowledge economy, intellectual property has become a key element of a company's core competitiveness. Currently, existing enterprise information query platforms and patent search systems on the market can, to a certain extent, link enterprise information with its intellectual property rights, listing the intellectual property rights held by the company and providing users with basic data integration services.

[0003] For companies with complex structures, their intellectual property is not concentrated in a single entity, but rather dispersed across multiple entities such as the parent company, subsidiaries, and affiliated companies. Since collaborative R&D among these entities is a key driver of technological development, understanding a company's collaborative R&D model is crucial for investment due diligence and risk assessment.

[0004] However, existing technologies are relatively weak in deeply analyzing the relationship between corporate R&D and intellectual property. For example, while the retrieved information may show that both the parent company and its subsidiaries have patents, it cannot directly demonstrate the technological connection between the subsidiaries and the parent company. This makes it difficult for users to determine the center and hub of technological R&D, affecting the assessment of the collaborative efficiency of a company's R&D system. Summary of the Invention

[0005] The purpose of this invention is to provide a method for processing intellectual property data, and the technical problem to be solved is how to analyze the correlation of collaborative R&D among enterprises based on intellectual property.

[0006] This invention is achieved through the following technical solution:

[0007] A method for processing intellectual property data includes the following steps:

[0008] Acquire intellectual property data and clean the aforementioned intellectual property data;

[0009] Extract the objectives and keywords of each intellectual property document from the cleaned intellectual property data; associate the objectives, keywords and intellectual property documents to generate an objective-intellectual property map.

[0010] When the target of the aforementioned intellectual property document includes at least two applicants, the aforementioned applicants shall be regarded as collaborating entities, and the intellectual property document shall be regarded as a collaborating document.

[0011] From the above target-intellectual property map, obtain the keywords of each intellectual property document of the collaborating entities and determine the contribution area of ​​each collaborating entity on the collaborative document;

[0012] The number of intellectual property documents of the aforementioned collaborating entities is counted to determine the first degree of relevance among the collaborating entities.

[0013] By generating a target-IP map, related elements of intellectual property are displayed in a structured and visualized manner, making enterprise IP information more organized and visible. Based on this, collaborative entities and documents are further identified, along with the contribution areas of each entity. This visually presents the relationships and technical contributions of different entities in collaborative R&D, overcoming the limitation of existing technologies that cannot intuitively demonstrate technical relationships. Using the number of IP documents as a metric to determine the primary degree of correlation between collaborative entities quantifies the closeness of R&D connections between them, helping users identify the centers and hubs of technology development and providing more valuable reference for investment due diligence and risk assessment.

[0014] Furthermore, the aforementioned targets include multiple subject types, each subject type including at least one subject; each of the aforementioned subjects is assigned a unique identity code, which is used to associate targets, keywords, and intellectual property documents.

[0015] To solve the problem of entities having the same name, each entity is assigned a unique identification code to prevent association with the wrong entity.

[0016] Furthermore, keywords for each intellectual property document from the aforementioned target-IP map are obtained, and the contribution areas of each collaborating entity on the collaborative document are determined, including the following steps:

[0017] The aforementioned intellectual property documents are categorized into contribution areas based on keywords;

[0018] Obtain the keywords from the above collaborative documents to obtain collaborative keywords;

[0019] Based on the aforementioned collaborative entities, keywords associated with these collaborative entities are obtained, resulting in a keyword library tagged with the collaborative entities.

[0020] Based on the above collaborative keywords, several synonyms are generated, and synonyms of the collaborative keywords are obtained from the above keyword library;

[0021] If the above synonyms are obtained, mark the corresponding collaborative subject on the collaborative keyword;

[0022] If the above synonyms are not obtained, then determine the second degree of relevance between the collaborative keywords and each collaborative subject;

[0023] The collaborative entities with the highest second degree of relevance mentioned above are selected as the markers for collaborative keywords;

[0024] Obtain the contribution area corresponding to the above collaborative keywords, and determine the contribution area of ​​the collaborative subject in the collaborative document based on the collaborative subject marked by the above collaborative keywords.

[0025] The system generates synonyms and retrieves them from a keyword database, taking into full account the technical concepts related to collaborative keywords. This avoids the omission of collaborative subject contribution information due to different keyword expressions, and improves the comprehensiveness and accuracy of the contribution analysis of collaborative subjects in collaborative manuscripts.

[0026] For unobtained collaborative keywords, the most relevant collaborative entity is determined by calculating the second degree of relevance. Finally, the contribution area of ​​each collaborative entity in the collaborative document is determined based on the collaborative keyword tags, so as to intuitively display the specific contribution of each collaborative entity in the collaborative R&D results. This allows users to more clearly understand the roles and functions of different applicant entities in technology R&D, further solving the problem that existing technologies cannot intuitively derive the technical relationship, and helps to analyze the relevance of enterprise collaborative R&D in a more in-depth way.

[0027] Furthermore, determining the second degree of relevance between collaborative keywords and each collaborative entity includes the following steps:

[0028] By statistically analyzing the frequency of each keyword in the aforementioned keyword database, the keyword with the highest frequency is selected as the main keyword for that collaborative entity.

[0029] The above main keywords and collaborative keywords are vectorized to obtain the main feature vector and collaborative feature vector;

[0030] The similarity between the aforementioned collaborative feature vector and the main feature vector is calculated to obtain the second correlation degree.

[0031] By analyzing keyword frequency to select primary keywords, and then calculating their similarity to collaborative keywords, the degree of relevance between the collaborating entity and the collaborating keywords in terms of core concerns and technical directions is reflected. This quantification of the relevance between the collaborating keywords and the collaborating entity makes the assessment of relevance more objective and avoids errors caused by subjective judgment.

[0032] Furthermore, the aforementioned collaborative entities include a first collaborative entity and a second collaborative entity;

[0033] Determining the first degree of association between the first and second collaborative entities includes the following steps:

[0034] The total number of intellectual property documents of the aforementioned first collaborative entity is counted;

[0035] The above-mentioned objectives include the total number of collaborative documents from the first and second collaborative entities;

[0036] The first degree of relevance between the first collaborative entity and the second collaborative entity is determined by the total number of intellectual property documents and the total number of collaborative documents of the first collaborative entity.

[0037] The calculation of the first degree of correlation from the perspective of the first collaborating entity is only valid for the analysis of the first collaborating entity.

[0038] Furthermore, record each time point and corresponding legal status of the aforementioned intellectual property documents to obtain a record ledger; link the aforementioned record ledger with the intellectual property documents.

[0039] The legal status at each point in time can be checked through the record ledger.

[0040] Furthermore, upon receiving a query instruction, the input target is obtained, and based on the aforementioned target-IP map query input target, keywords, IP documents, and record ledgers are obtained.

[0041] Furthermore, after retrieving the aforementioned intellectual property documents, the number of intellectual property documents was counted.

[0042] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0043] By generating a target-IP map, related elements of intellectual property are displayed in a structured and visualized manner, making enterprise IP information more organized and visible. Based on this, collaborative entities and documents are further identified, along with the contribution areas of each entity. This visually presents the relationships and technical contributions of different entities in collaborative R&D, overcoming the limitation of existing technologies that cannot intuitively demonstrate technical relationships. Using the number of IP documents as a metric to determine the primary degree of correlation between collaborative entities quantifies the closeness of R&D connections between them, helping users identify the centers and hubs of technology development and providing more valuable reference for investment due diligence and risk assessment. Attached Figure Description

[0044] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be considered as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort. In the drawings:

[0045] Figure 1 Main flowchart. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0047] First embodiment:

[0048] Combination Figure 1 A method for processing intellectual property data includes the following steps:

[0049] Acquire intellectual property data, including but not limited to trademarks, patents, and software copyrights, and clean the aforementioned intellectual property data; reduce the impact of non-standard data on the results of collaborative R&D correlation analysis.

[0050] The objectives and keywords of each intellectual property document are extracted from the cleaned intellectual property data; the objectives, keywords and intellectual property documents are linked to generate an objective-intellectual property map; this map constructs the relationship between intellectual property-related elements, which helps to grasp intellectual property information as a whole.

[0051] When the target of the aforementioned intellectual property document includes at least two applicants, the applicants are considered as collaborating entities, and the intellectual property document is considered as a collaborating document. Defining the applicants for collaborative research and development and the intellectual property documents involved provides specific targets for subsequent analysis of the relationship between collaborating entities.

[0052] By extracting keywords from each intellectual property document of the collaborating entities in the aforementioned target-intellectual property map, the contribution areas of each collaborating entity on the collaborative document can be determined. By identifying the contribution area of ​​each collaborating entity through keywords, the specific contribution of each collaborating entity in the collaborative R&D results (intellectual property document) can be seen intuitively, which helps to gain a deeper understanding of the role and function of each applicant entity in technology R&D.

[0053] By counting the number of intellectual property documents of the aforementioned collaborating entities, the primary degree of relevance between them can be determined. Using the number of intellectual property documents as a metric to determine relevance quantifies the closeness of R&D collaboration among the collaborating entities, providing a basis for evaluating the collaborative efficiency of an enterprise's R&D system.

[0054] By generating a target-IP map, related elements of intellectual property are displayed in a structured and visualized manner, making enterprise IP information more organized and visible. Based on this, collaborative entities and documents are further identified, along with the contribution areas of each entity. This visually presents the relationships and technical contributions of different entities in collaborative R&D, overcoming the limitation of existing technologies that cannot intuitively demonstrate technical relationships. Using the number of IP documents as a metric to determine the primary degree of correlation between collaborative entities quantifies the closeness of R&D connections between them, helping users identify the centers and hubs of technology development and providing more valuable reference for investment due diligence and risk assessment.

[0055] In a specific embodiment, the above-mentioned targets include multiple subject types, and each subject type includes at least one subject; each of the above-mentioned subjects is assigned a unique identity code, which is used to associate the target, keywords and intellectual property documents.

[0056] To solve the problem of entities having the same name, each entity is assigned a unique identification code to prevent association with the wrong entity.

[0057] In a specific implementation, each time point and corresponding legal status of the aforementioned intellectual property document is recorded to obtain a record ledger; this record ledger is then linked to the intellectual property document. The legal status of each time point can be queried through the record ledger.

[0058] In a specific implementation, upon receiving a query command, the input target is obtained. Based on the aforementioned target-IP map, the input target is queried to obtain keywords, IP documents, and record ledgers. After retrieving the aforementioned IP documents, the number of IP documents is counted.

[0059] This is a reference-grade trademark data set that uses regular expressions or rules to process non-standard data (such as addresses and date formats), employs OCR technology to recognize text information on trademark images, extracts targets (applicant, patentee, inventor, address, etc.), and records each time point in the trademark application process and its corresponding legal status (acceptance, examination, authorization, rejection, etc.) for subsequent queries. Specifically: it allows queries based on notification letters to find corresponding time points and legal statuses; queries based on legal status to find corresponding time points; and queries based on time points or legal status to find nodes in the application process.

[0060] The identified applicants, patentees, and inventors are associated with each other. Applicants and patentees can be either companies or individuals. When an applicant or patentee is a company, the company's business registration information is associated with it. This information includes the company name, registered address, and social credit code.

[0061] The analysis and statistics of trademarks are performed on a company-by-company basis. When any item in the business registration information matches the target information, the corresponding target, keywords, intellectual property documents, and records are extracted. If any other input target does not match the extracted target, the entity of that target is marked, and the number of marks is counted. Specifically: the applicant of trademark a is company A, address B, and identity code C; the applicant of trademark b is company A, address B', and identity code C; the applicant of trademark c is company A, address B, and identity code C'; and the applicant of trademark d is company E, address B', and identity code C'. Company A's registered address is known to be B, and its social credit code is C. The input target for the query is company A, and trademarks a to c are retrieved. Since the address B' of trademark b is inconsistent with the registered address B of company A, the address of trademark b is marked as an address error, and the number of address error marks is counted. Since the identity code C' of trademark c is inconsistent with the social credit code C of company A, the identity code of trademark c is marked as a coding error, and the number of coding error marks is counted. The number of patents can also be counted based on patent category or current legal status.

[0062] Second embodiment:

[0063] Based on the first embodiment, keywords for each intellectual property document of the collaborating entities are obtained from the aforementioned target-intellectual property map, and the contribution areas of each collaborating entity on the collaborative documents are determined, including the following steps:

[0064] The aforementioned intellectual property documents are categorized into contribution areas based on keywords;

[0065] Obtain the keywords from the above collaborative documents to obtain collaborative keywords;

[0066] Based on the aforementioned collaborative entities, keywords associated with these collaborative entities are obtained, resulting in a keyword library tagged with the collaborative entities.

[0067] Based on the aforementioned collaborative keywords, several synonyms can be generated. An upper limit can be set for the number of generated synonyms, thus solving the problem of reduced data processing efficiency caused by the unlimited expansion of synonyms. Synonyms for the collaborative keywords are obtained from the aforementioned keyword library.

[0068] If the above synonyms are obtained, the corresponding collaborating entity is marked on the collaborative keyword; multiple marks may exist for the collaborative keyword at the same time, that is, when multiple collaborating entities can complete this part of the content, the contribution value of their contribution area can be distributed equally.

[0069] If the above synonyms are not obtained, then determine the second degree of relevance between the collaborative keywords and each collaborative subject;

[0070] The collaborative entities with the highest second degree of relevance mentioned above are selected as the markers for collaborative keywords;

[0071] Obtain the contribution area corresponding to the above collaborative keywords, and determine the contribution area of ​​the collaborative subject in the collaborative document based on the collaborative subject marked by the above collaborative keywords.

[0072] The system generates synonyms and retrieves them from a keyword database, taking into full account the technical concepts related to collaborative keywords. This avoids the omission of collaborative subject contribution information due to different keyword expressions, and improves the comprehensiveness and accuracy of the contribution analysis of collaborative subjects in collaborative manuscripts.

[0073] For unobtained collaborative keywords, the most relevant collaborative entity is determined by calculating the second degree of relevance. Finally, the contribution area of ​​each collaborative entity in the collaborative document is determined based on the collaborative keyword tags, so as to intuitively display the specific contribution of each collaborative entity in the collaborative R&D results. This allows users to more clearly understand the roles and functions of different applicant entities in technology R&D, further solving the problem that existing technologies cannot intuitively derive the technical relationship, and helps to analyze the relevance of enterprise collaborative R&D in a more in-depth way.

[0074] Third embodiment:

[0075] Based on the second embodiment, the second degree of association between collaborative keywords and each collaborative subject is determined, including the following steps:

[0076] By statistically analyzing the frequency of each keyword in the aforementioned keyword database, the keyword with the highest frequency is selected as the main keyword for that collaborative entity.

[0077] The above-mentioned main keywords and co-keywords can be vectorized using existing text-shifting models (such as Jina Embeddings v2 model, bag-of-words model, etc.) to obtain main feature vectors and co-feature vectors.

[0078] The similarity between the aforementioned collaborative feature vector and the main feature vector can be calculated using the cosine similarity algorithm to obtain the second correlation degree.

[0079] By analyzing keyword frequency to select primary keywords, and then calculating their similarity to collaborative keywords, the degree of relevance between the collaborating entity and the collaborating keywords in terms of core concerns and technical directions is reflected. This quantification of the relevance between the collaborating keywords and the collaborating entity makes the assessment of relevance more objective and avoids errors caused by subjective judgment.

[0080] Fourth embodiment:

[0081] Based on any of the above embodiments, each of the above collaborative entities includes a first collaborative entity and a second collaborative entity;

[0082] Determining the first degree of association between the first and second collaborative entities includes the following steps:

[0083] The total number of intellectual property documents of the aforementioned first collaborative entity is counted;

[0084] The above-mentioned objectives include the total number of collaborative documents from the first and second collaborative entities;

[0085] The first degree of relevance between the first collaborative entity and the second collaborative entity is determined by the total number of intellectual property documents and the total number of collaborative documents of the first collaborative entity.

[0086] The calculation of the first degree of correlation from the perspective of the first collaborating entity is only valid for the analysis of the first collaborating entity.

[0087] If Company A applies for 100 patents and Company B applies for 1000 patents, with Company A and Company B as joint applicants for 20 patents and Company A, Company B, and Company C as joint applicants for 10 patents; when determining the degree of collaboration between Company A and Company B in R&D, from Company A's perspective, through... The calculated first degree of association between company A and company B is: ;in, Indicates the first degree of relevance; Indicates the total number of intellectual property documents; This indicates the total number of collaborative documents. From Company B's perspective... The calculated first degree of association between company B and company A is: .

[0088] In addition, a company's business registration information can be linked to other dimensions of data, including but not limited to qualifications, honors, business information, and risk information. Intellectual property data and business registration information can be stored in a document database (MongoDB), which can store unstructured text and is easily expandable.

[0089] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for processing intellectual property data, characterized in that, Includes the following steps: Obtain intellectual property data and clean the intellectual property data; Extract the target and keywords for each intellectual property document from the cleaned intellectual property data; By associating the stated objectives, keywords, and intellectual property documents, a target-intellectual property map is generated. When the target of the intellectual property document includes at least two applicants, the applicants are regarded as collaborating entities, and the intellectual property document is regarded as a collaborating document. Keywords for each intellectual property document of the collaborating entities are obtained from the target-intellectual property map, and the contribution areas of each collaborating entity on the collaborative document are determined. The number of intellectual property documents of the collaborating entities is counted to determine the first degree of association between the collaborating entities.

2. The processing method according to claim 1, characterized in that, The targets include multiple subject types, and each subject type includes at least one subject; each subject is assigned a unique identity code, which is used to associate the target, keywords and intellectual property documents.

3. The processing method according to claim 1, characterized in that, The process of extracting keywords from each intellectual property document of the collaborating entities from the target-intellectual property map and determining the contribution areas of each collaborating entity on the collaborative documents includes the following steps: The intellectual property manuscripts are divided into contribution areas according to keywords; Obtain the keywords from the collaborative document to obtain the collaborative keywords; Based on the collaborative entity, keywords associated with the collaborative entity are obtained to obtain a keyword library marked with the collaborative entity; Several synonyms are generated based on the collaborative keywords, and synonyms of the collaborative keywords are obtained from the keyword library; If the synonym is obtained, the corresponding collaborative entity is marked on the collaborative keyword; If the synonyms are not obtained, then determine the second degree of association between the collaborative keywords and each collaborative subject; Select the collaborating entity with the highest second degree of relevance as the marker for the collaborating keyword; Obtain the contribution area corresponding to the collaborative keyword, and determine the contribution area of ​​the collaborative subject in the collaborative document based on the collaborative subject marked by the collaborative keyword.

4. The processing method according to claim 3, characterized in that, Determining the secondary relevance between collaborative keywords and each collaborative entity includes the following steps: The frequency of each keyword of the collaborating entity is counted by the keyword database, and the keyword with the highest frequency is selected as the main keyword of the collaborating entity. The main keywords and collaborative keywords are vectorized to obtain the main feature vector and collaborative feature vector; The similarity between the collaborative feature vector and the main feature vector is calculated to obtain the second correlation degree.

5. The processing method according to claim 1, characterized in that, Each collaborative entity includes a first collaborative entity and a second collaborative entity; Determining the first degree of association between the first and second collaborative entities includes the following steps: The total number of intellectual property documents of the first collaborative entity is counted; The statistical target includes the total number of collaborative documents from the first and second collaborative entities; The first degree of association between the first collaborative entity and the second collaborative entity is determined by the total number of intellectual property documents and the total number of collaborative documents of the first collaborative entity.

6. The processing method according to claim 1, characterized in that, Record each time point and corresponding legal status of the intellectual property document to obtain a record ledger; link the record ledger with the intellectual property document.

7. The processing method according to claim 6, characterized in that, Upon receiving a query instruction, the input target is obtained, and the input target is queried based on the target-intellectual property map to obtain keywords, intellectual property documents, and record ledgers.

8. The processing method according to claim 7, characterized in that, After retrieving the intellectual property documents, the number of intellectual property documents was counted.

Images