Research And Implementation Of Economic Crime Investigation Intelligence Analysis System Based On Knowledge Graph

The rapid development of the Internet and modern information technology have not only changed the way people live and produce,but also provided criminals with more means and models of crime,especially in the economic field.The criminal activities in economic cases are gradually integrated with the Internet,which makes the clues of the cases and the evidence of the crimes easily drowned by the intricate network of relationships,posing a huge challenge to the intelligence analysis of investigation.In recent years,with the further development of artificial intelligence technology,the knowledge graph has set off a wave of research.As a semantic web,knowledge graph has powerful semantic processing capabilities and open data organization capabilities,and is gradually being used in various fields.This paper analyzes the current problems of intelligence analysis in economic crime investigation and the characteristics of knowledge graph technology,researches and implements an economic crime investigation intelligence analysis system based on knowledge graph to promote the application of knowledge graph in the field of economic crime investigation,and then promote the transformation and innovation of economic crime investigation intelligence analysis.The work done in this article mainly includes the following points:Ⅰ.Representation,modeling,and storage of the knowledge graph of economic crime investigation intelligence.By comparing the characteristics of mainstream knowledge graph representation methods and storage technologies,and combining the needs of economic intelligence information models and knowledge storage requirements,this paper proposes a top-down economic intelligence knowledge modeling method based on RDF and RDFs and a knowledge storage method of economic crime investigation intelligence based on MongoDB.Ⅱ.Knowledge extraction of economic crime investigation intelligence.In terms of knowledge extraction,this paper first studies the method of knowledge extraction for structured data,and uses it to extract some intelligence knowledge from structured enterprise basic information,personnel information and other data to form a knowledge graph;then applies the LightGBM model in machine learning to identify the shell companies in the knowledge base and complete the corresponding label knowledge;finally,based on the existing knowledge,according to RDFs reasoning rules and custom domain reasoning rules,Jena reasoning machine is used for knowledge reasoning to further complete knowledge graph.Ⅲ.Visualization and automatic layout of economic crime investigation intelligence knowledge graph.This paper studies the development history of Web visualization technology,the advantages and disadvantages of several common layout algorithms,and their scopes of application.Based on the requirements of graph visualization in this paper,GPU accelerated client rendering technologies are selected to perform graph node and edge Visualization,combined with JavaScript to achieve some graph-based human-computer interaction,and finally a force-oriented layout model based on the server is designed to optimize the layout of the graph.Ⅳ.Based on the research of the above three parts and the requirements of the knowledge graph-based economic intelligence analysis system,the three major functional modules of the system—knowledge data management module,graphic visualization module and intelligence analysis module were designed and implemented.