CN111274332A - Intelligent patent retrieval method and system based on knowledge graph - Google Patents
Intelligent patent retrieval method and system based on knowledge graph Download PDFInfo
- Publication number
- CN111274332A CN111274332A CN202010061494.XA CN202010061494A CN111274332A CN 111274332 A CN111274332 A CN 111274332A CN 202010061494 A CN202010061494 A CN 202010061494A CN 111274332 A CN111274332 A CN 111274332A
- Authority
- CN
- China
- Prior art keywords
- retrieval
- information
- patents
- similarity
- knowledge graph
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/288—Entity relationship models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/3332—Query translation
- G06F16/3334—Selection or weighting of terms from queries, including natural language queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Animal Behavior & Ethology (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a patent intelligent retrieval method and system based on a knowledge graph, which are characterized in that original information of each patent is obtained from a database, basic information, a patent author and a patent application unit in the original information are used as entities, entity nodes of each patent and the author of each patent are constructed through attribute information of the entities and the entities, and a knowledge graph library containing all entity nodes and relations among the entity nodes is obtained according to the relations among the entity nodes; acquiring retrieval information, judging whether the retrieval information is basic information, if so, retrieving all patents with similarity degrees larger than a preset value corresponding to the retrieval information in a knowledge graph library to serve as retrieval results, otherwise, judging whether the retrieval information is attribute information, if so, describing the attribute information through a graph-like relation to generate retrieval subgraphs, performing subgraph matching in the knowledge graph library according to the retrieval subgraphs to obtain a matching result set, taking the patents with similarity values larger than the preset value as retrieval results, and otherwise, ending the retrieval.
Description
Technical Field
The invention relates to the fields of information technology, artificial intelligence and intellectual property, in particular to an intellectual map-based patent intelligent retrieval method and system.
Background
By 2018, China has ranked the first worldwide in patent application number for 8 consecutive years. Under such circumstances, how to obtain effective patent information from a large amount of patent data is an important basis for further knowledge innovation. The current patent retrieval method still adopts a retrieval mode of keyword matching and full-text retrieval core. The retrieval method is easy to have the problems of large matching amount of retrieval contents, low matching precision and large amount of secondary retrieval required by a user, so that the user cannot acquire accurate patent data in a short time. In recent years, with the extensive research of graph data technology, a method for organizing, searching and pushing knowledge information of vertical industries based on a knowledge graph has been applied to a certain extent. When the knowledge graph is used for data discovery, the amount of the interview is considered except for the keywords, so that a more accurate retrieval result can be provided. Based on the consideration, the patent designed by the application is to adopt a knowledge graph method to construct an intelligent patent retrieval method.
The main technical background of the patent quality evaluation method designed by the application is as follows:
1) graph database systems have become one of the major database systems, providing a solid data storage foundation for the construction and retrieval of knowledge maps.
The knowledge graph is a network structure system based on the relationship between elements, and in order to realize the storage and the retrieval of the knowledge graph data, the knowledge graph data must be stored by adopting a graph. Matching and retrieving graph data has been a difficult problem. In recent years, with the intensive research, a large-scale database system represented by Neo4j is in a stage of commercial application, and provides good support for storage, matching and retrieval of graph data. And also provides a solid application foundation for the smooth application of the knowledge graph. The knowledge-graph storage and construction employed in the present application is based on graph data.
2) The machine learning is applied in a large scale, and the technical foundation for automatically constructing the knowledge graph is provided.
The construction of the knowledge graph relates to the automatic extraction of a large amount of data to realize the acquisition of the relationship between the node entities. Meanwhile, the weight among different relations in the knowledge graph plays a vital role in the accuracy of the knowledge graph, the weight needs to be automatically adjusted through the using process of the knowledge graph, and the wide application of machine learning in recent years provides important technical support for the automatic adjustment of the knowledge graph.
In conclusion, the knowledge graph of the application provides an intelligent patent retrieval method.
Disclosure of Invention
The patent intelligent retrieval method is provided based on knowledge graph. The main inventive content related to the application comprises:
1) a method for constructing a patent knowledge graph. The method constructs a graph description mode of the patent based on different elements forming the patent, and constructs a knowledge graph of the patent according to the data description mode. Meanwhile, an automatic construction process of the patent knowledge graph is given;
2) the method for constructing the knowledge graph query structure by utilizing the user input is characterized in that a retrieval element used for inputting is constructed into a query description which can be understood by a knowledge graph library;
3) a patent retrieval method based on a patent knowledge graph.
Specifically speaking, the invention provides an intelligent patent retrieval method based on knowledge graph, which comprises the following steps:
step 1, acquiring original information of each patent from a database, taking basic information, a patent author and a patent application unit in the original information as entities, constructing entity nodes of each patent and the author thereof through attribute information of the entities and the entities, and acquiring a knowledge graph library containing all the entity nodes and relations among the entity nodes according to the relations among the entity nodes;
and 2, acquiring retrieval information, judging whether the retrieval information is basic information, if so, retrieving all patents with similarity greater than a preset value corresponding to the retrieval information in the knowledge graph library to serve as retrieval results, otherwise, judging whether the retrieval information is attribute information, if so, describing the attribute information through a graph-like relation, generating retrieval sub-graphs, performing sub-graph matching in the knowledge graph library according to the retrieval sub-graphs to obtain a matching result set, taking the patents with similarity greater than the preset value as retrieval results, and otherwise, ending the retrieval.
The intelligent patent retrieval method based on the knowledge graph comprises the following steps of: similarity between patents and citation relationship between patents.
The intelligent patent retrieval method based on the knowledge graph comprises the following steps:
step 11, acquiring attribute sets S and S 'of any two patents p and p' respectively;
step 12, for any S ∈ S and S ' ∈ S ', if S ∈ ═ S ' and S is an IPC classification number or a master classification number, step 14 is entered, and otherwise step 13 is entered;
step 13, setting the similarity of the patents p and p' as 0;
step 14, obtaining the keyword set in the classification fieldThe keyword set is utilized to carry out word segmentation on full texts of patents p and p' respectively, and word frequency is calculated, so that word segmentation vectors K ═ K { (K) are obtained respectively1,k2,.,knK'1,k′2,k′m}。
And step 15, calculating the similarity between K and K 'by using a similarity algorithm to serve as the similarity between the patents p and p'.
The intelligent retrieval method for the patent based on the knowledge graph comprises the following steps of: keywords, the inventor and the technical field.
The intelligent retrieval method for the patent based on the knowledge graph is characterized in that the similarity algorithm is a text similarity calculation algorithm.
The invention also provides a patent intelligent retrieval system based on the knowledge graph, which comprises the following steps:
the method comprises the steps that a module 1 acquires original information of each patent from a database, takes basic information, a patent author and a patent application unit in the original information as entities, constructs entity nodes of each patent and the author thereof through attribute information of the entities and the entities, and acquires a knowledge graph library containing all the entity nodes and relations among the entity nodes according to the relations among the entity nodes;
and a module 2, acquiring retrieval information, judging whether the retrieval information is basic information, if so, retrieving all patents with similarity greater than a preset value corresponding to the retrieval information in the knowledge graph library as retrieval results, otherwise, judging whether the retrieval information is attribute information, if so, describing the attribute information through a graph-like relationship, generating retrieval sub-graphs, performing sub-graph matching in the knowledge graph library according to the retrieval sub-graphs to obtain a matching result set, taking the patents with similarity greater than the preset value as retrieval results, and otherwise, ending the retrieval.
The intellectual retrieval system of patent based on knowledge graph, wherein the relation among the entity nodes includes: similarity between patents and citation relationship between patents.
The intelligent patent retrieval system based on the knowledge graph comprises a similarity calculation system and a similarity calculation system, wherein the similarity calculation system comprises:
a module 11, for any two patents p and p ', respectively acquiring attribute sets s and s';
the module 12, for any S ∈ S and S ' ∈ S ', if S ∈ ═ S ' and S is an IPC classification number or a master classification number, then enter the module 14, otherwise enter the module 13;
setting the similarity of the patents p and p' as 0 by a module 13;
the module 14 obtains a keyword set κ in the classification field, performs word segmentation on the full texts of the patents p and p' by using the keyword set, and calculates word frequency to obtain a word segmentation vector K ═ { κ ═ respectively1,κ2,.κnK'1,k′2,k′m}。
And a module 15 for calculating the similarity between K and K 'by using a similarity algorithm to serve as the similarity between the patents p and p'.
The intelligent retrieval system for the patent based on the knowledge graph comprises the following steps of: keywords, the inventor and the technical field.
The intelligent retrieval system for the patent based on the knowledge graph is characterized in that the similarity algorithm is a text similarity calculation algorithm.
According to the scheme, the invention has the advantages that: the method effectively describes the correlation among patents in a knowledge graph mode, realizes the retrieval among the related patent contents by a graph matching method, reduces the search space based on the traditional retrieval methods such as keywords and the like, and further provides the similarity among the related patents by utilizing similarity calculation.
Drawings
FIG. 1 is an overall framework of the present invention;
FIG. 2 is a basic composition diagram of a patent knowledge map;
FIG. 3 is a flow chart for constructing a patent knowledge graph;
figure 4 patent retrieval subgraph.
Detailed Description
In order to make the aforementioned features and effects of the present invention more comprehensible, embodiments accompanied with figures are described in detail below.
The overall framework. The general framework of patent intelligent retrieval is shown in fig. 1, and is divided into 3 steps, including:
and a step of constructing a patent knowledge graph, wherein the patent knowledge graph is constructed by utilizing relevant information of patents to construct interrelations among the patents, attributes and patents. And (3) constructing a patent knowledge graph according to the step 2.
And a step of generating a patent retrieval subgraph, wherein the process of generating the patent retrieval subgraph is to generate a relationship description subgraph of the patent retrieval through a graph relationship description method according to the input of a user for retrieval in a knowledge graph database.
And a knowledge graph retrieval step, wherein the knowledge graph is retrieved by completing management retrieval in the knowledge graph according to the patent retrieval subgraph, and outputting a retrieval result. The process of retrieving the knowledge-graph is seen in step 3.
And (5) constructing a patent knowledge graph.
2-1. composition and construction process of knowledge graph:
a patent knowledge graph is a relational network that describes the relationships between patents and patent attributes and between patents. The basic description of the patent knowledge map is shown in fig. 2. A circle in the patent knowledge map identifies an entity, and a square identifies the attribute of the entity. In the knowledge graph of the patent, the entities include the following types: patents, patent authors and patent application units. For each entity, there is a set of attributes that describe the set of intrinsic attributes of the entity. For the above three entities, the attributes considered are as follows:
the patent attributes are as follows: application number, authorization number, validity period, keyword, abstract, major classification number, publication date, IPC classification number;
attributes of the patent author: author name, author unit, author address;
attributes of patent application units: unit name, unit property, unit address.
Entity nodes of a patent and an author thereof can be constructed through entity and attribute information. After the nodes are constructed, the relationships between patents and authors and between patents are further constructed. The relationship between patents and authors identifies the authors of patents, and includes two categories: 1) similarly, identifying similarity values between patents, and referring to 2-2 for a similarity calculation method of the patents; 2) the citation relationship between patents, such as pre-patents, identifies whether patent a is a pre-patent of patent B, including priority based on patent a or sub-patents based on patent a.
The process of constructing the whole patent knowledge map library is shown in fig. 3.
2-2 patent similarity calculation
The patent similarity calculation is to calculate the similarity degree between any two patents, and takes the text similarity calculation as the core. Any text similarity calculation method can be adopted for calculation, and the text similarity calculation method is beyond the coverage range of the application and is not described in detail. However, the text similarity algorithm is calculated by using text keywords as feature vectors, so that if no distinction is made, similarity calculation is performed on any two patents, and a similarity value exists. It is clear that for patents in different fields there is no need for similarity calculation. Therefore, the similarity calculation for the patent must be filtered. In summary, the process of patent similarity calculation is as follows:
step 1), acquiring attribute sets s and s 'of any two patents p and p' respectively;
and step 2), for any S ∈ S and S '∈ S', S 'is a corresponding attribute set of the patent p'. Because two patents p and p ' are adopted, two different symbols are adopted to correspond to different attribute sets of the two patents p and p ', if s is equal to s ' and s is an IPC classification number or a main classification number, the step 4) is carried out, and otherwise, the step 3) is carried out;
step 3), setting the similarity of the patents p and p' as 0;
and 4), acquiring a keyword set kappa in the classification field, segmenting full texts of patents p and p' by using the keyword set, and calculating word frequency to obtain a segmentation vector K ═ { K ═ respectively1,k2,.,knK'1,k′2,k′m}。
And 5) calculating the similarity between K and K 'by using a similarity algorithm, namely the similarity between the patents p and p'.
And (5) patent retrieval process. The general flow of patent retrieval is as follows:
step 1) user inputs search information I
Step 2), if I is a single keyword, ending
Step 3), if the I is a patent, acquiring all other patents with similarity greater than v corresponding to the patent I, and returning a result, wherein v is called a similarity threshold value, the initial value is 70%, and the initial value can be adjusted by a user;
step 4), if I is a set of patent description attribute sets, the attribute sets need to be satisfiedWherein, keyword is abstract or keyword in keyword, Author is Author, domain is technical field. Then for the attribute set, a patent retrieval sub-graph g is constructed, and for constructing a patent sub-graph, a description language of the constructed sub-graph of the graph database system can be used. For example, the Neo4j database, which has the Cypher language (similar to SQL language) standard to describe a query subgraph. The subgraph example is only given to show the construction mode of the subgraph, and the specific subgraph description is different from different databases, so that the method of the invention is suitable for various databases. The style of the patent retrieval subgraph is shown in FIG. 4 (the meaning shown in FIG. 4 is that the retrieval target is all patents X with attributes 1, 2 and 3 and author A);
step 5), carrying out sub-graph matching in the knowledge graph library according to the query sub-graph g to obtain a matching result set Ro;
Step 6), for any r eoAcquiring all patent sets R with all satisfied similarity values larger than vrThe similarity calculation can adopt text similarity calculation, namely, the text similarity calculation is carried out on the patent search results matched through the subgraphs for further filtering.
Step 7), returning the search result set R to ∪ Rr(r∈Ro)。
In the process, the steps 5-6 are used for taking the subset of the attribute set as a further search target, and filtering the similarity of the set, so that a wider result can be obtained.
The following are system examples corresponding to the above method examples, and this embodiment can be implemented in cooperation with the above embodiments. The related technical details mentioned in the above embodiments are still valid in this embodiment, and are not described herein again in order to reduce repetition. Accordingly, the related-art details mentioned in the present embodiment can also be applied to the above-described embodiments.
The invention also provides a patent intelligent retrieval system based on the knowledge graph, which comprises the following steps:
the method comprises the steps that a module 1 acquires original information of each patent from a database, takes basic information, a patent author and a patent application unit in the original information as entities, constructs entity nodes of each patent and the author thereof through attribute information of the entities and the entities, and acquires a knowledge graph library containing all the entity nodes and relations among the entity nodes according to the relations among the entity nodes;
and a module 2, acquiring retrieval information, judging whether the retrieval information is basic information, if so, retrieving all patents with similarity greater than a preset value corresponding to the retrieval information in the knowledge graph library as retrieval results, otherwise, judging whether the retrieval information is attribute information, if so, describing the attribute information through a graph-like relationship, generating retrieval sub-graphs, performing sub-graph matching in the knowledge graph library according to the retrieval sub-graphs to obtain a matching result set, taking the patents with similarity greater than the preset value as retrieval results, and otherwise, ending the retrieval.
The intellectual retrieval system of patent based on knowledge graph, wherein the relation among the entity nodes includes: similarity between patents and citation relationship between patents.
The intelligent patent retrieval system based on the knowledge graph comprises a similarity calculation system and a similarity calculation system, wherein the similarity calculation system comprises:
a module 11, for any two patents p and p ', respectively acquiring attribute sets S and S';
the module 12, for any S ∈ S and S ' ∈ S ', if S ═ S ' and S is an IPC classification number or a master classification number, then enter the module 14, otherwise enter the module 13;
setting the similarity of the patents p and p' as 0 by a module 13;
module 14 for obtaining a set k of keywords in the classification domain, using the relationThe key word set respectively carries out word segmentation on full texts of patents p and p' and calculates word frequency, so that word segmentation vectors K-K are respectively obtained1,k2,.,knK'1,k′2,k′m}。
And a module 15 for calculating the similarity between K and K 'by using a similarity algorithm to serve as the similarity between the patents p and p'.
The intelligent retrieval system for the patent based on the knowledge graph comprises the following steps of: keywords, the inventor and the technical field.
The intelligent retrieval system for the patent based on the knowledge graph is characterized in that the similarity algorithm is a text similarity calculation algorithm.
Claims (10)
1. An intelligent patent retrieval method based on knowledge graph is characterized by comprising the following steps:
step 1, acquiring original information of each patent from a database, taking basic information, a patent author and a patent application unit in the original information as entities, constructing entity nodes of each patent and the author thereof through attribute information of the entities and the entities, and acquiring a knowledge graph library containing all the entity nodes and relations among the entity nodes according to the relations among the entity nodes;
and 2, acquiring retrieval information, judging whether the retrieval information is basic information, if so, retrieving all patents with similarity greater than a preset value corresponding to the retrieval information in the knowledge graph library to serve as retrieval results, otherwise, judging whether the retrieval information is attribute information, if so, describing the attribute information through a graph-like relation, generating retrieval sub-graphs, performing sub-graph matching in the knowledge graph library according to the retrieval sub-graphs to obtain a matching result set, taking the patents with similarity greater than the preset value as retrieval results, and otherwise, ending the retrieval.
2. The intellectual retrieval system patent based on knowledge graph as claimed in claim 1, wherein the relationship between the entity nodes comprises: similarity between patents and citation relationship between patents.
3. The intellectual retrieval system patent based on knowledge graph as claimed in claim 2, wherein the similarity calculation method is:
step 11, acquiring attribute sets S and S 'of any two patents p and p' respectively;
step 12, for any S ∈ S and S ' ∈ S ', if S ∈ ═ S ' and S is an IPC classification number or a master classification number, step 14 is entered, and otherwise step 13 is entered;
step 13, setting the similarity of the patents p and p' as 0;
step 14, obtaining a keyword set κ in the classification field, segmenting the full texts of the patents p and p' respectively by using the keyword set, and calculating word frequency, thereby obtaining a segmentation vector K ═ { K ═ respectively1,k2,.,knK'1,k′2,k′m}。
And step 15, calculating the similarity between K and K 'by using a similarity algorithm to serve as the similarity between the patents p and p'.
4. The intellectual property retrieval method based on the knowledge-graph as claimed in claim 1, wherein the attribute information of the retrieval information includes: keywords, the inventor and the technical field.
5. The intellectual retrieval system patent based on knowledge-graph as claimed in claim 3, wherein the similarity algorithm is a text similarity calculation algorithm.
6. An intelligent patent retrieval system based on knowledge graph is characterized by comprising:
the method comprises the steps that a module 1 acquires original information of each patent from a database, takes basic information, a patent author and a patent application unit in the original information as entities, constructs entity nodes of each patent and the author thereof through attribute information of the entities and the entities, and acquires a knowledge graph library containing all the entity nodes and relations among the entity nodes according to the relations among the entity nodes;
and a module 2, acquiring retrieval information, judging whether the retrieval information is basic information, if so, retrieving all patents with similarity greater than a preset value corresponding to the retrieval information in the knowledge graph library as retrieval results, otherwise, judging whether the retrieval information is attribute information, if so, describing the attribute information through a graph-like relationship, generating retrieval sub-graphs, performing sub-graph matching in the knowledge graph library according to the retrieval sub-graphs to obtain a matching result set, taking the patents with similarity greater than the preset value as retrieval results, and otherwise, ending the retrieval.
7. The intellectual retrieval system based on knowledge-graph as claimed in claim 6, wherein the relationship between the entity nodes comprises: similarity between patents and citation relationship between patents.
8. The intellectual property knowledge graph based patent intelligent retrieval system as claimed in claim 7, wherein the similarity calculation system is:
a module 11, for any two patents p and p ', respectively acquiring attribute sets s and s';
the module 12, for any S ∈ S and S ' ∈ S ', if S ∈ ═ S ' and S is an IPC classification number or a master classification number, then enter the module 14, otherwise enter the module 13;
setting the similarity of the patents p and p' as 0 by a module 13;
the module 14 obtains a keyword set κ in the classification field, performs word segmentation on the full texts of the patents p and p' by using the keyword set, and calculates word frequency to obtain a word segmentation vector K ═ K ″, respectively1,k2,.,knK'1,k′2,k′m}。
And a module 15 for calculating the similarity between K and K 'by using a similarity algorithm to serve as the similarity between the patents p and p'.
9. The intellectual property graph based patent intelligent retrieval system as claimed in claim 6, wherein the attribute information of the retrieval information comprises: keywords, the inventor and the technical field.
10. The intellectual property map based patent intelligent retrieval system as claimed in claim 8 wherein the similarity algorithm is a text similarity calculation algorithm.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010061494.XA CN111274332A (en) | 2020-01-19 | 2020-01-19 | Intelligent patent retrieval method and system based on knowledge graph |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010061494.XA CN111274332A (en) | 2020-01-19 | 2020-01-19 | Intelligent patent retrieval method and system based on knowledge graph |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111274332A true CN111274332A (en) | 2020-06-12 |
Family
ID=70997488
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010061494.XA Pending CN111274332A (en) | 2020-01-19 | 2020-01-19 | Intelligent patent retrieval method and system based on knowledge graph |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111274332A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112306687A (en) * | 2020-10-30 | 2021-02-02 | 平安数字信息科技(深圳)有限公司 | Resource allocation method and device based on knowledge graph, computer equipment and medium |
CN113190692A (en) * | 2021-05-28 | 2021-07-30 | 山东顺势教育科技有限公司 | Self-adaptive retrieval method, system and device for knowledge graph |
CN113255324A (en) * | 2021-03-09 | 2021-08-13 | 西安循数信息科技有限公司 | Method for disambiguating inventor names in patent data |
CN113868294A (en) * | 2021-08-31 | 2021-12-31 | 北京中知智慧科技有限公司 | Intellectual property retrieval method and device based on explosion diagram |
CN117251539A (en) * | 2023-08-11 | 2023-12-19 | 北京中知智慧科技有限公司 | Patent intelligent retrieval system using generative artificial intelligence |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109189942A (en) * | 2018-09-12 | 2019-01-11 | 山东大学 | A kind of construction method and device of patent data knowledge mapping |
CN109522465A (en) * | 2018-10-22 | 2019-03-26 | 国家电网公司 | The semantic searching method and device of knowledge based map |
CN110399457A (en) * | 2019-07-01 | 2019-11-01 | 吉林大学 | A kind of intelligent answer method and system |
CN110516047A (en) * | 2019-09-02 | 2019-11-29 | 湖南工业大学 | The search method and searching system of knowledge mapping based on packaging field |
-
2020
- 2020-01-19 CN CN202010061494.XA patent/CN111274332A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109189942A (en) * | 2018-09-12 | 2019-01-11 | 山东大学 | A kind of construction method and device of patent data knowledge mapping |
CN109522465A (en) * | 2018-10-22 | 2019-03-26 | 国家电网公司 | The semantic searching method and device of knowledge based map |
CN110399457A (en) * | 2019-07-01 | 2019-11-01 | 吉林大学 | A kind of intelligent answer method and system |
CN110516047A (en) * | 2019-09-02 | 2019-11-29 | 湖南工业大学 | The search method and searching system of knowledge mapping based on packaging field |
Non-Patent Citations (2)
Title |
---|
阮彤;王梦婕;王昊奋;胡芳槐;: "垂直知识图谱的构建与应用研究" * |
陈锡瑞: "基于知识图谱的情报关联分析方法研究" * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112306687A (en) * | 2020-10-30 | 2021-02-02 | 平安数字信息科技(深圳)有限公司 | Resource allocation method and device based on knowledge graph, computer equipment and medium |
CN113255324A (en) * | 2021-03-09 | 2021-08-13 | 西安循数信息科技有限公司 | Method for disambiguating inventor names in patent data |
CN113255324B (en) * | 2021-03-09 | 2022-02-18 | 西安循数信息科技有限公司 | Method for disambiguating inventor names in patent data |
CN113190692A (en) * | 2021-05-28 | 2021-07-30 | 山东顺势教育科技有限公司 | Self-adaptive retrieval method, system and device for knowledge graph |
CN113190692B (en) * | 2021-05-28 | 2022-06-24 | 山东顺势教育科技有限公司 | Self-adaptive retrieval method, system and device for knowledge graph |
CN113868294A (en) * | 2021-08-31 | 2021-12-31 | 北京中知智慧科技有限公司 | Intellectual property retrieval method and device based on explosion diagram |
CN117251539A (en) * | 2023-08-11 | 2023-12-19 | 北京中知智慧科技有限公司 | Patent intelligent retrieval system using generative artificial intelligence |
CN117251539B (en) * | 2023-08-11 | 2024-04-02 | 北京中知智慧科技有限公司 | Patent intelligent retrieval system using generative artificial intelligence |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111274332A (en) | Intelligent patent retrieval method and system based on knowledge graph | |
CN104239513B (en) | A kind of semantic retrieving method of domain-oriented data | |
CN111966793B (en) | Intelligent question-answering method and system based on knowledge graph and knowledge graph updating system | |
CN107193882B (en) | Why-not query answer method based on graph matching on RDF data | |
CN111382276B (en) | Event development context graph generation method | |
WO2004013774A2 (en) | Search engine for non-textual data | |
WO2004013772A2 (en) | System and method for indexing non-textual data | |
WO2004013775A2 (en) | Data search system and method using mutual subsethood measures | |
CN109508385B (en) | Character relation analysis method in webpage news data based on Bayesian network | |
CN109241278A (en) | Scientific research knowledge management method and system | |
CN115563313A (en) | Knowledge graph-based document book semantic retrieval system | |
CN103761286B (en) | A kind of Service Source search method based on user interest | |
Saad et al. | Efficient skyline computation on uncertain dimensions | |
Paton et al. | Dataset discovery and exploration: A survey | |
CN113761124A (en) | Training method of text coding model, information retrieval method and equipment | |
Yan et al. | The self-adaptive integrated algorithm on time-sensitive cluster evolution tracking | |
CN117435685A (en) | Document retrieval method, document retrieval device, computer equipment, storage medium and product | |
CN117010373A (en) | Recommendation method for category and group to which asset management data of power equipment belong | |
CN108536796B (en) | Heterogeneous ontology matching method and system based on graph | |
CN116244497A (en) | Cross-domain paper recommendation method based on heterogeneous data embedding | |
Mohemad et al. | Ontological-based information extraction of construction tender documents | |
Gao et al. | Association and Recomendation for Geosciences Data Attributes Based on Semantic Similarity Measurement | |
Pushpalatha et al. | A tree based representation for effective pattern discovery from multimedia documents | |
Giglou et al. | ParsSimpleQA: The Persian Simple Question Answering Dataset and System over Knowledge Graph | |
Sotiropoulos et al. | A genetic algorithm approach for topic clustering: A centroid-based encoding scheme |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |