CN-122019783-A - Case knowledge graph construction method, electronic equipment and storage medium

CN122019783ACN 122019783 ACN122019783 ACN 122019783ACN-122019783-A

Abstract

The application provides a case knowledge graph construction method, electronic equipment and a storage medium, and relates to the technical field of big data and artificial intelligence application. And extracting the preprocessed case file text through an information extraction model to obtain entity information. And mapping the entity information to obtain a graph node, and constructing a relation edge according to the graph node and the information in the case file text. And constructing a case knowledge graph based on the graph nodes and the relation edges. The application introduces a framework combining natural language processing and knowledge graph, realizes the structured extraction of the information of each entity in the case file text, and builds a graph database to model the relationship among the entities. The method breaks the limitation of the traditional information island and provides a new idea for identifying the upstream and downstream role chains.

Inventors

MA YULONG
QI YUTING
HE YIXIN
XIONG WEIJIA
FU HUIYUAN

Assignees

北京邮电大学

Dates

Publication Date: 20260512
Application Date: 20251113
Priority Date: 20250818

Claims (10)

1. The case knowledge graph construction method is characterized by comprising the following steps of: acquiring a case file text, and preprocessing the case file text; Extracting the preprocessed case file text through an information extraction model to obtain entity information; mapping to obtain a graph node based on the entity information, and constructing a relation edge according to the graph node and the information in the case file text; And constructing a case knowledge graph based on the graph nodes and the relation edges.
2. The method of claim 1, wherein the preprocessing the case volume text comprises: carrying out data cleaning and unified format operation on the file text of the case; removing messy code characters in the case file text, and uniformly coding the case file text; and carrying out paragraph segmentation and key field indexing on the case file text.
3. The method of claim 1, wherein extracting the preprocessed case volume text by the information extraction model to obtain the entity information comprises: Identifying personnel identity information and case information in the preprocessed case file text; extracting a communication account number in the preprocessed case file text through a regular expression; Invoking a financial account number in the preprocessed case file text, and checking the authenticity of the financial account number; Identifying legal person information in the preprocessed case file text; And taking the identity information, the case information, the communication account number, the financial account number and the legal person information as the entity information.
4. A method according to claim 3, wherein said mapping based on said entity information to obtain a graph node comprises: constructing suspicious personnel nodes based on the identity information and the case information; Constructing suspicious account nodes according to the communication account, the financial account and other financial information in the preprocessed case file text; constructing enterprise legal person nodes based on the legal person information; And taking the suspicious personnel node, the suspicious account node and the enterprise legal person node as the graph nodes.
5. The method of claim 1, wherein constructing a relationship edge from the graph node and information in case volume text comprises: And constructing a fund circulation relation side, a communication relation side and a personnel subordinate relation side among the graph nodes according to the information in the case file text, wherein the attribute of the fund circulation relation side at least comprises transaction time, the attribute of the communication relation side at least comprises communication time, and the attribute of the personnel subordinate relation side at least comprises a role.
6. The method as recited in claim 1, further comprising: Calculating the weight value of the relation edge through the following formula: Weight value=ln (transaction amount) ×communication frequency coefficient.
7. The method as recited in claim 1, further comprising: responding to the acquired new case file text, and extracting a new communication account number and a new financial account number in the new case file text; And based on the new communication account number and the new financial account number, matching is carried out in the case knowledge graph, and the fund flow direction between the associated cases is analyzed and calculated in response to matching to the associated cases.
8. The method as recited in claim 1, further comprising: and visually displaying the case knowledge graph.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and running on the processor, characterized in that the processor implements the method according to any one of claims 1 to 8 when executing the computer program.
10. A non-transitory computer readable storage medium storing computer instructions for causing a computer to perform the method of any one of claims 1 to 8.

Description

Case knowledge graph construction method, electronic equipment and storage medium Technical Field The application relates to the technical field of big data and artificial intelligence application, in particular to a case knowledge graph construction method, electronic equipment and a storage medium. Background And the problems of information dispersion, low association mining efficiency and the like exist in the handling of the associated cases. The related cases lack structural integration, the upstream and downstream clues are distributed in a fragmented manner, and the base layer business still depends on a primary mode of 'manual file browsing+basic database retrieval'. The association mining is low-efficiency, the existing digital tool is only limited to matching single information elements, and the deep semantic analysis and dynamic relation mining capability is lacked, so that the complex case processing period is as long as 6-12 months, and the resources are in long-term overload operation. Disclosure of Invention In view of the above, the present application aims to provide a case knowledge graph construction method, an electronic device and a storage medium, so as to solve the problem of low associated case mining efficiency. Based on the above object, a first aspect of the present application provides a case knowledge graph construction method, including: acquiring a case file text, and preprocessing the case file text; Extracting the preprocessed case file text through an information extraction model to obtain entity information; mapping to obtain a graph node based on the entity information, and constructing a relation edge according to the graph node and the information in the case file text; And constructing a case knowledge graph based on the graph nodes and the relation edges. Optionally, the preprocessing the case file text includes: carrying out data cleaning and unified format operation on the file text of the case; removing messy code characters in the case file text, and uniformly coding the case file text; and carrying out paragraph segmentation and key field indexing on the case file text. Optionally, the extracting, by the information extraction model, the preprocessed case file text to obtain entity information includes: Identifying personnel identity information and case information in the preprocessed case file text; extracting a communication account number in the preprocessed case file text through a regular expression; Invoking a financial account number in the preprocessed case file text, and checking the authenticity of the financial account number; Identifying legal person information in the preprocessed case file text; And taking the identity information, the case information, the communication account number, the financial account number and the legal person information as the entity information. Optionally, the mapping to obtain the graph node based on the entity information includes: constructing suspicious personnel nodes based on the identity information and the case information; Constructing suspicious account nodes according to the communication account, the financial account and other financial information in the preprocessed case file text; constructing enterprise legal person nodes based on the legal person information; And taking the suspicious personnel node, the suspicious account node and the enterprise legal person node as the graph nodes. Optionally, the constructing a relationship edge according to the graph node and the information in the case file text includes: And constructing a fund circulation relation side, a communication relation side and a personnel subordinate relation side among the graph nodes according to the information in the case file text, wherein the attribute of the fund circulation relation side at least comprises transaction time, the attribute of the communication relation side at least comprises communication time, and the attribute of the personnel subordinate relation side at least comprises a role. Optionally, the method further comprises: Calculating the weight value of the relation edge through the following formula: Weight value=ln (transaction amount) ×communication frequency coefficient. Optionally, the method further comprises: responding to the acquired new case file text, and extracting a new communication account number and a new financial account number in the new case file text; And based on the new communication account number and the new financial account number, matching is carried out in the case knowledge graph, and the fund flow direction between the associated cases is analyzed and calculated in response to matching to the associated cases. Optionally, the method further comprises the step of visually displaying the case knowledge graph. Based on the same inventive concept, a second aspect of the application also provides an electronic device comprising a memory, a processor and a computer program stored on the