Serial/incorporated case identification method

Abstract

The invention discloses a serial / incorporated case identification method and system. The method comprises the following steps of carrying out preprocessing: carrying out word segmentation and part-of-speech tagging on the case detail description of a case, and removing stop words; extracting the important characteristics of the case from the preprocessed case detail description through a method which combines a rule with a dictionary, and converting a case text into a characteristic vector; and according to the characteristic vector, adopting a case characteristic similarity calculation method to obtain the comprehensive similarity of the cases, clustering the cases through the comprehensive similarity, and finding the dense cluster of related cases to identify the serial / incorporated cases. Through the method and the system provided by the invention, clustering is carried out on the basis of characteristic density, and limitation that a traditional clustering algorithm is applied to a numeric type vector is eliminated. A difficult point that possible serial / incorporated case clusters cannot be obtained from a case detail text library is overcome. The method and the system can be applied to case investigation to improve investigation efficiency.

Description

technical field [0001] The invention belongs to the technical field of computer natural language processing and data mining, and more specifically relates to a method for identifying serial and parallel cases. Background technique [0002] As an important method of combating serial criminal cases, serial and parallel case analysis can tap the internal connection between cases, reduce the workload of analysts, and improve the efficiency of solving cases. The criminal case text contains information such as the time of the case, the location of the case, the modus operandi and tools, etc. Using this information combined with data mining methods to mine the internal links between cases and discover the clusters of cases can reduce the workload of analysts and improve the efficiency of solving cases. efficiency. [0003] The traditional techniques for discovering dense clusters mainly use clustering methods, among which, density-based clustering methods can discover clusters of ...

AI Technical Summary

Problems solved by technology

[0004] Although traditional clustering algorithms can find clusters of arbitrary shapes and densities, their application targets are mainly for points in numerical multidimensional vector spaces; criminal case texts are short and contain a lot of important information, but they are free texts without fixed formats. It is impossible to directly discover the association between cases through traditional clustering methods

Images