Entity data association method, system, device and medium based on knowledge graph

Abstract

The present invention provides an entity data association method, system, device and medium based on a knowledge graph, and relates to the field of computer application technology. The entity data association method includes: obtaining data to be entered into the graph, the data to be entered into the graph includes multiple initial fields; obtaining the target field that needs to be fuzzy matched in the multiple initial fields; creating an ES index containing the target field according to the data to be entered into the graph, and importing the data to be entered into the ES index after initialization; traversing each data in the data to be entered into the graph, fuzzily matching the target field in each data with the ES index to obtain fuzzy associated data; generating side information with an associated relationship according to the fuzzy associated data, and storing the fuzzy associated data and side information in a graph database. The present invention can identify and associate the fuzzy relationship between entity data, and establish a complete graph data association relationship.

Description

Technical Field

[0001] The present invention relates to the field of computer application technology, and more specifically, to a knowledge graph-based entity data association method, system, device and medium. Background Art

[0002] In the traditional graph relationship generation process, only points with exactly the same primary key attributes can be automatically connected. Points with different but similar primary key attributes cannot be automatically connected, so it is impossible to establish a complete graph data association relationship for non-standard data. For example, a graph database can automatically connect certain association relationships, such as customer A is associated with a mobile phone number 1, and customer B is also associated with the same mobile phone number 1. Because the mobile phone numbers associated with customer A and customer B are certain and completely consistent, the graph database will automatically form an association relationship of "customer A----mobile phone number 1----customer B" when the data is written. However, the graph database cannot directly identify and associate fuzzy relationships. For example, customer A is associated with a work unit Xingfu Consumer Finance, and customer B is associated with a work unit Xingfu Consumer Finance. From the perspective of social cognition, customer A and customer B are associated with the same company, but in the graph database, since the vertex data of Xingfu Consumer Finance and Xingfu Consumer Finance are not exactly the same, a connection relationship cannot be automatically formed.

[0003] With the continuous development of financial technology, knowledge graphs are being used more and more widely in the field of financial risk control. The point and edge data structure of the graph enables it to efficiently process complex relationship data and help anti-fraud systems identify customer relationship risks. However, the relationships in the graph are not very perfect and intelligent when they are formed, and the problems of the aforementioned traditional graph relationship generation methods still exist. In the field of big data risk control, the customer data collected is often not completely standard. How to use these non-standard data to establish a complete graph data association relationship has become a major pain point in the construction of knowledge graphs in the risk control field.

[0004] Therefore, there is an urgent need for a technical solution that can accurately identify the relationships between various entities in similar but not completely consistent data, so as to expand the scope of risk identification and provide protection for the construction and application of graph data. Summary of the invention

[0005] In order to solve the problems existing in the above-mentioned prior art or at least part of the problems, the embodiments of the present invention provide an entity data association method, system, storage medium, device and computer program product based on a knowledge graph. By fuzzy matching each piece of data to be entered into the graph with the ES index containing the target field to obtain fuzzy associated data with fuzzy association connections and then entering them into the graph, the fuzzy relationship between entity data can be identified and associated, and a complete graph data association relationship can be established.

[0006] According to the first aspect of the present invention, an embodiment of the present invention provides an entity data association method based on a knowledge graph, which includes: obtaining data to be entered into the graph, wherein the data to be entered into the graph includes multiple initial fields; obtaining a target field that needs to be fuzzy matched among the multiple initial fields; creating an ES index containing the target field based on the data to be entered into the graph, and importing the data to be entered into the ES index after initializing it; traversing each data in the data to be entered into the graph, and fuzzily matching the target field in each data with the ES index to obtain fuzzy associated data; generating side information with an associated relationship based on the fuzzy associated data, and storing the fuzzy associated data and the side information in a graph database.

[0007] According to the above-mentioned embodiment of the present invention, by fuzzy matching each piece of data to be entered into the graph with the ES index containing the target field to obtain fuzzy associated data with fuzzy associated connections, and then generating side information with associated relationships based on the fuzzy associated data and re-entering the graph, the fuzzy relationships between entity data can be identified and associated, thereby establishing a complete graph data association relationship for non-standard data.

[0008] In some embodiments of the present invention, traversing each piece of data in the data to be entered into the graph, and fuzzy matching the target field in each piece of data with the ES index to obtain fuzzy associated data includes: fuzzy matching the target field of a first piece of data in the data to be entered into the graph with an index to be matched in the ES index to obtain a corresponding similarity score; when the similarity score is higher than a preset score value, the fuzzy associated data includes the first data and the second data corresponding to the index to be matched.

[0009] In some embodiments of the present invention, the fuzzy association data further includes the similarity score.

[0010] In some embodiments of the present invention, traversing each piece of data in the data to be entered into the graph, fuzzy matching the target field in each piece of data with the ES index to obtain fuzzy related data also includes: when the similarity score is higher than a preset score value, deleting the first data and the second data in the data to be entered into the graph.

[0011] According to the second aspect of the present invention, an embodiment of the present invention provides an entity data association system based on a knowledge graph, which includes: an initial data acquisition module, used to acquire data to be entered into the graph, wherein the data to be entered into the graph includes multiple initial fields; an ES index generation module, used to acquire a target field that needs to be fuzzy matched among the multiple initial fields, create an ES index containing the target field according to the data to be entered into the graph, and import the data to be entered into the ES index after initialization; an associated data acquisition module, used to traverse each data in the data to be entered into the graph, fuzzily match the target field in each data with the ES index to obtain fuzzy associated data; a knowledge graph generation module, used to generate side information with an associated relationship according to the fuzzy associated data, and store the fuzzy associated data and side information in a graph database.

[0012] According to the above-mentioned embodiment of the present invention, by fuzzy matching each piece of data to be entered into the graph with the ES index containing the target field to obtain fuzzy associated data with fuzzy associated connections, and then generating side information with associated relationships based on the fuzzy associated data and re-entering the graph, the fuzzy relationships between entity data can be identified and associated, thereby establishing a complete graph data association relationship for non-standard data.

[0013] In some embodiments of the present invention, traversing each piece of data in the data to be entered into the graph, and fuzzy matching the target field in each piece of data with the ES index to obtain fuzzy associated data includes: fuzzy matching the target field of a first piece of data in the data to be entered into the graph with an index to be matched in the ES index to obtain a corresponding similarity score; when the similarity score is higher than a preset score value, the fuzzy associated data includes the first data and the second data corresponding to the index to be matched.

[0014] In some embodiments of the present invention, the fuzzy association data further includes the similarity score.

[0015] In some embodiments of the present invention, traversing each piece of data in the data to be entered into the graph, fuzzy matching the target field in each piece of data with the ES index to obtain fuzzy related data also includes: when the similarity score is higher than a preset score value, deleting the first data and the second data in the data to be entered into the graph.

[0016] According to the third aspect of the present invention, an embodiment of the present invention provides a computer-readable storage medium having computer-readable instructions stored thereon. When the computer-readable instructions are executed by a processor, the computer performs the following operations: the operations include the steps included in the entity data association method described in any of the above embodiments.

[0017] According to a fourth aspect of the present invention, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory is used to store one or more computer-readable instructions, wherein the one or more computer-readable instructions, when executed by the processor, can implement the entity data association method described in any one of the above embodiments.

[0018] According to a fifth aspect of the present invention, an embodiment of the present invention provides a computer program product including a computer program, and when the computer program is executed by a processor, the entity data association method as described in any one of the above embodiments is implemented.

[0019] From the above, it can be seen that the entity data association method, system, storage medium, device and computer program product provided by the embodiments of the present invention can identify and associate the fuzzy relationship between entity data by fuzzily matching each piece of data to be entered into the graph with the ES index containing the target field to obtain fuzzy associated data with fuzzy associated connections, and then generate side information with associated relationships based on the fuzzy associated data and re-enter the graph, thereby establishing a complete graph data association relationship for non-standard data. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a schematic diagram of a flow chart of a method for associating entity data based on a knowledge graph according to Embodiment 1 of the present invention;

[0021] Figure 2 is a schematic diagram of the connection relationship of associated data in a graph database according to an embodiment of the present invention;

[0022] Figure 3 It is a schematic diagram of the architecture of an entity data association system based on a knowledge graph according to Example 2 of the present invention. DETAILED DESCRIPTION

[0023] The various aspects of the present invention are described in detail below in conjunction with the accompanying drawings and specific embodiments. Among them, well-known modules, units and their connections, links, communications or operations are not shown or described in detail. In addition, the described features, architectures or functions can be combined in any way in one or more embodiments. It should be understood by those skilled in the art that the various embodiments described below are only used for illustration and are not intended to limit the scope of protection of the present invention. It can also be easily understood that the modules or units or processing methods in each embodiment described herein and shown in the accompanying drawings can be combined and designed in various different configurations.

[0024] The following is a brief description of the terms used in this article.

[0025] ES: A distributed, RESTful, full-text search engine based on the Lucene library that can quickly store, search, and analyze large amounts of data and supports multiple data types and complex query operations.

[0026] [Example 1]

[0027] Figure 1 It is a flowchart of the entity data association method based on the knowledge graph according to Example 1 of the present invention.

[0028] like Figure 1 As shown, in Embodiment 1 of the present invention, the entity data association method may include: step S11, step S12, step S13, step S14 and step S15, and the above steps are described in detail below.

[0029] In step S11, the data to be input into the image is obtained, and the data to be input into the image includes a plurality of initial fields.

[0030] In some implementations, the data to be entered into the graph is a table existing in a relational database, such as a customer resume table, which includes two initial fields: "customer, work unit".

[0031] In this embodiment, before step S11, the normal deployment and startup of ES and graph database services are completed through the deployment of relevant components and the preparation of the test environment.

[0032] In step S12, a target field that needs to be fuzzy matched is obtained from the multiple initial fields, wherein the target field is at least one field that needs to be fuzzy matched, for example, if the work unit of a customer needs to be fuzzily associated in the graph database, the target field is "work unit".

[0033] In step S13, an ES index including the target field is created according to the data to be entered, and the data to be entered is initialized and imported into the ES index, wherein the target field in the ES index is of Text type.

[0034] For example, when it is necessary to fuzzily associate the work unit of a customer in a graph database, a customer resume index is created in ES, and the index fields are "customer, work unit", where the "work unit" index field type is Text type to support word segmentation; after the ES index is created, the relationship table data is imported to complete the initialization operation of the existing data, and the data initialized in the ES index is the determined range of data to be entered into the graph.

[0035] In step S14, each piece of data in the data to be entered into the graph is traversed, and the target field in each piece of data is fuzzily matched with the ES index to obtain fuzzy associated data.

[0036] In some embodiments, traversing each piece of data in the data to be entered into the graph, and fuzzily matching the target field in each piece of data with the ES index to obtain fuzzy associated data includes: fuzzily matching the target field of a first piece of data in the data to be entered into the graph with an index to be matched in the ES index to obtain a corresponding similarity score, and when the similarity score is higher than a preset score value, the fuzzy associated data includes the first data and the second data corresponding to the index to be matched. Optionally, the fuzzy associated data also includes the similarity score between the first data and the second data that have a fuzzy associated relationship.

[0037] In a further embodiment, traversing each piece of data in the data to be entered into the graph, fuzzy matching the target field in each piece of data with the ES index to obtain fuzzy related data also includes: when the similarity score is higher than a preset score value, deleting the first data and the second data in the data to be entered into the graph.

[0038] In step S15, side information with association relationship is generated according to the fuzzy association data, and the fuzzy association data and the side information are stored in a graph database.

[0039] In an exemplary embodiment, for the customer resume to be added to the map, the following steps are performed to add the resume to the map:

[0040] (1) Traverse the customer resume data in the relational database.

[0041] (2) Perform fuzzy matching on the ES customer resume index based on the work unit, and sort the matching results by similarity score (for example, sort from high to low by similarity score). The ES query result has a relevance score field.

[0042] _score, the higher the relevance score, the higher the similarity.

[0043] (3) Determine whether the ES query result has hit related data based on the access rules. If hit related data, the group of customer data (including the input customers and the hit customers) is saved.

[0044] The records are fuzzy related data with fuzzy related relationships, and this group of data is removed from the relational customer resume table. For example, the admission rule is: _score>80 means that the two work units are related, then the ES query result _score higher than 80 means hit, there is a similar related work unit; _score lower than 80 means miss, there is no similar related work unit.

[0045] (4) Continue to traverse the remaining customer data in the relational customer resume table and repeat step (3) until all data are matched.

[0046] (5) Obtain the customer association data (i.e., fuzzy association data) recorded in step (3), convert each set of related customer data into a corresponding point-edge structure, and batch import them into the graph. For example, in step (3), it is recorded that customer C's unit A1 (Xingfu Consumer Finance) is similar to customer D's unit A2 (Xingfu Consumer Finance), and the similarity score is 85. Then, it is converted into a point-edge structure of "customer C----unit A1--similarity: 85--unit A2----customer D", and the following is generated in the graph database: Figure 2 The connection relationship of the input graph is shown. Optional, Figure 2 The circles in are represented as points in the figure.

[0047] In a further embodiment, the entity data association method further includes: acquiring historical data in a graph database; and performing fuzzy matching on an ES index generated according to the data to be entered into the graph based on the historical data to acquire fuzzy associated data.

[0048] By using the above-mentioned entity data association method based on knowledge graph of Example 1 of the present invention, by combining ES and graph database, ES can be used to realize the characteristics of data fuzzy matching, instead of using ES as a secondary index to speed up the query speed of graph data. Before entering the graph, the data can be first established through ES to establish a fuzzy association relationship between the new data and the historical data before entering the graph. The characteristics of the full-text search engine can be used to make up for the disadvantage that the graph database cannot automatically identify and associate fuzzy relationships, thereby enriching the association relationship between entity data. Applying the above-mentioned entity data association method of Example 1 of the present invention to the field of financial risk control can enrich the association relationship between the entities of risk control data, effectively expand the scope of risk identification, and provide more data support for anti-fraud.

[0049] In a further embodiment, after the graph in the graph database is generated according to the entity data association method based on the knowledge graph of the first embodiment of the invention, the data association relationship is obtained by querying the graph database. For example, taking the graph connection relationship generated in step (5) in the above graph entry example as an example, by exploring the edge relationship of the vertex of customer C, it can be found that the work units of customer C and customer D are related, and the two are suspected to be in the same work unit.

[0050] [Example 2]

[0051] Figure 3 It is a schematic diagram of the architecture of an entity data association system based on a knowledge graph according to Example 2 of the present invention.

[0052] like Figure 3As shown, the entity data association system includes: an initial data acquisition module 310, an ES index generation module 320, an associated data acquisition module 330, and a knowledge graph generation module 340.

[0053] The initial data acquisition module 310 is used to acquire the data to be entered into the image, and the data to be entered into the image includes a plurality of initial fields.

[0054] The ES index generation module 320 is used to obtain a target field that needs to be fuzzy matched from the multiple initial fields, create an ES index containing the target field according to the data to be entered, and import the data to be entered into the ES index after initialization.

[0055] The associated data acquisition module 330 is used to traverse each piece of data in the data to be entered into the graph, and perform fuzzy matching between the target field in each piece of data and the ES index to obtain fuzzy associated data.

[0056] In some embodiments, traversing each piece of data in the data to be entered into the graph, and fuzzily matching the target field in each piece of data with the ES index to obtain fuzzy associated data includes: fuzzily matching the target field of a first piece of data in the data to be entered into the graph with an index to be matched in the ES index to obtain a corresponding similarity score, and when the similarity score is higher than a preset score value, the fuzzy associated data includes the first data and the second data corresponding to the index to be matched. Optionally, the fuzzy associated data also includes the similarity score between the first data and the second data that have a fuzzy associated relationship.

[0057] In a further embodiment, traversing each piece of data in the data to be entered into the graph, fuzzy matching the target field in each piece of data with the ES index to obtain fuzzy related data also includes: when the similarity score is higher than a preset score value, deleting the first data and the second data in the data to be entered into the graph.

[0058] In a further embodiment, the associated data acquisition module is further used to execute the following steps: acquiring historical data in the graph database; and performing fuzzy matching on the ES index generated according to the data to be entered into the graph based on the historical data to acquire fuzzy associated data.

[0059] The knowledge graph generation module 340 is used to generate side information with association relationships based on the fuzzy association data, and store the fuzzy association data and side information in a graph database.

[0060] The above-mentioned entity data association system based on knowledge graph of Example 2 of the present invention is adopted. By fuzzy matching each data in the data to be entered into the graph with the ES index containing the target field, fuzzy associated data with fuzzy associated connections are obtained, and then side information with associated relationships is generated based on the fuzzy associated data and re-entered into the graph. It can identify and associate the fuzzy relationships between entity data, thereby establishing a complete graph data association relationship for non-standard data.

[0061] Through the description of the above implementation modes, those skilled in the art can clearly understand that the present invention can be implemented by combining software with a hardware platform. Based on such an understanding, all or part of the contribution of the technical solution of the present invention to the background technology can be embodied in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various implementation modes of the present invention or some parts of the implementation modes.

[0062] Correspondingly, an embodiment of the present invention further provides a computer-readable storage medium on which computer-readable instructions or programs are stored. When the computer-readable instructions or programs are executed by a processor, the computer performs the following operations: the operations include the steps included in the entity data association method described in any of the above embodiments, which are not repeated here. The storage medium may include, for example, a CD, a hard disk, a floppy disk, a flash memory, a magnetic tape, etc.

[0063] In addition, an embodiment of the present invention further provides a computer device including a memory and a processor, wherein the memory is used to store one or more computer-readable instructions or programs, wherein the one or more computer-readable instructions or programs can implement the entity data association method described in any of the above embodiments when executed by the processor. The computer device can be, for example, a server, a desktop computer, a notebook computer, a tablet computer, etc.

[0064] The present invention also provides a computer program product including a computer program, wherein the computer program includes program code for executing the entity data association method shown in the flowchart. When the computer program product is run in a computer system, the program code is used to enable the computer system to implement the entity data association method provided by the embodiment of the present disclosure.

[0065] According to an embodiment of the present disclosure, the program code for executing the computer program provided by the embodiment of the present disclosure can be written in any combination of one or more programming languages. Specifically, these computing programs can be implemented using high-level process and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, Java, C++, python, "C" language or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on the remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect through the Internet).

[0066] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention is described in detail with reference to the above embodiments, a person skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention. Therefore, the protection scope of the present invention shall be subject to the claims.

Claims

1. A method for associating entity data based on knowledge graph, characterized in that: The entity data association method comprises: Acquire the data to be entered into the image, wherein the data to be entered into the image includes a plurality of initial fields; Obtaining a target field that needs to be fuzzy matched from the multiple initial fields; Create an ES index containing the target field according to the data to be entered into the graph, and import the data to be entered into the ES index after initialization; Traverse each piece of data in the data to be entered into the graph, and perform fuzzy matching between the target field in each piece of data and the ES index to obtain fuzzy related data; Generate side information with association relationship according to the fuzzy association data, and store the fuzzy association data and the side information in a graph database.

2. The entity data association method according to claim 1, characterized in that: Traversing each piece of data in the data to be entered into the graph, and fuzzily matching the target field in each piece of data with the ES index to obtain fuzzy related data includes: A target field of a first data in the data to be entered into the graph is fuzzily matched with an index to be matched in the ES index to obtain a corresponding similarity score. When the similarity score is higher than a preset score value, the fuzzy associated data includes the first data and the second data corresponding to the index to be matched.

3. The entity data association method according to claim 2, characterized in that: The fuzzy association data also includes the similarity score.

4. The entity data association method according to claim 2, characterized in that: Traversing each piece of data in the data to be entered into the graph, and fuzzily matching the target field in each piece of data with the ES index to obtain fuzzy associated data further includes: When the similarity score is higher than a preset score value, the first data and the second data in the data to be entered into the image are deleted.

5. An entity data association system based on knowledge graph, characterized in that: The entity data association system comprises: An initial data acquisition module, used to acquire data to be entered into a graph, wherein the data to be entered into a graph includes a plurality of initial fields; An ES index generation module is used to obtain a target field that needs to be fuzzy matched from the multiple initial fields, create an ES index containing the target field according to the data to be entered into the graph, and import the data to be entered into the ES index after initialization; A related data acquisition module, used for traversing each piece of data in the data to be entered into the graph, and performing fuzzy matching between the target field in each piece of data and the ES index to obtain fuzzy related data; The knowledge graph generation module is used to generate side information with association relationships based on the fuzzy association data, and store the fuzzy association data and side information in a graph database.

6. The entity data association system according to claim 5, characterized in that: Traversing each piece of data in the data to be entered into the graph, and fuzzily matching the target field in each piece of data with the ES index to obtain fuzzy related data includes: A target field of a first data in the data to be entered into the graph is fuzzily matched with an index to be matched in the ES index to obtain a corresponding similarity score. When the similarity score is higher than a preset score value, the fuzzy associated data includes the first data and the second data corresponding to the index to be matched.

7. The entity data association system according to claim 6, characterized in that: Traversing each piece of data in the data to be entered into the graph, and fuzzily matching the target field in each piece of data with the ES index to obtain fuzzy associated data further includes: When the similarity score is higher than a preset score value, the first data and the second data in the data to be entered into the image are deleted.

8. A computer-readable storage medium storing computer-readable instructions, characterized in that: The computer-readable instructions are executed by a processor to implement the entity data association method according to any one of claims 1 to 4.

9. A computer device comprising a memory and a processor, The memory stores computer-readable instructions, characterized in that: The processor executes the computer-readable instructions to implement the entity data association method according to any one of claims 1 to 4.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the entity data association method according to any one of claims 1 to 4 is implemented.

Images