Research And Application Of Key Technologies Of Discipline Inspection And Supervision Judgment Based On Multidimensional Big Data Analysis

Under the call of "Implementing the national big data strategy," the discipline inspection and supervision agencies are required to reshape the governance ecology.Aiming at the problems of data type disorder,difficult management,and difficulty finding clues in the processing of case investigation under the mode of "big data + discipline inspection and supervision," the thesis proposes to construct a case investigation preliminary knowledge graph based on the knowledge graph.By reasoning about its implicit relationship,analyzing and mining case patterns,and providing a data analysis visualization system,it can provide case investigators with useful preliminary clues and help investigations.This thesis has the following contributions:1.Propose a technology based on knowledge graph construction to deal with the initial nuclear,big data of multi-dimensional case investigation.In the process of knowledge extraction,the multiple types of Excel table data obtained by the discipline inspection and supervisionrelated information departments are extracted,merged,and converted into a triple form.In knowledge reasoning,the automatic reasoning machine is combined with the custom rule reasoning machine abstracted based on the past case investigation experience of the case-handling personnel,which overcomes the problem of incomplete triple reasoning in the implicit relationship in the automatic reasoning graph.The goal of effectively expanding the initial nuclear knowledge map of cases has been achieved.2.Propose the use of Knowledge Graph Construction technology to handle multi-dimensional case investigation big data.Through the use of abnormal information extraction methods to analyze the persons involved,the use of pattern judgment methods to mine crime patterns(especially gang patterns),and the use of the "knowledge map + community discovery" model to classify the persons involved,achieve the goal of providing case investigators with effective classification data and making case investigation more convenient.3.Design and implement a data and analysis result visualization system for case investigators.The analyzed data will be visualized from multiple perspectives,such as displaying time information of significant events from the perspective of the time link and displaying the itinerary information of the involved person of the space link.Besides,according to specific functional requirements,the system uses the corresponding Cypher statement to query the case investigation’s initial Knowledge Graph.It then visualizes the query results in the form of graphs and data on the front-end page.Finally,it provides a reasonable and simple visual platform for case handling personnel.The thesis analyzes the results of the construction,mining and analysis,and visualization of the preliminary knowledge Graph of disciplinary inspection and supervision case investigations,and indicates that the processing and analysis of multi-dimensional disciplinary inspection and supervision case investigation preliminary inspection data based on Knowledge Graph related technologies can dig out the hidden related information,abnormal information,and group information provide effective clues for case investigators and make case investigation more convenient.