Method and device for determining group members

Abstract

The invention provides a method and a device for determining group members. The method comprises the following steps: acquiring relationship data between people in a predetermined data set according to a first predetermined rule; determining an associated person set corresponding to each selected person according to one or more selected persons and the relationship data; determining members of a group according to the selected person and the associated person set corresponding to the selected person, wherein the first predetermined rule is a rule determined according to relationship characteristics among group members. According to the technical scheme, the members of the group are determined according to the characteristics of the group, and the group members can be determined more accurately.

Description

technical field [0001] The invention relates to the computer field, in particular to a method and device for determining group members. Background technique [0002] Compared with single-person crimes, gang crimes are more harmful and have a greater social impact. Through cooperation among gang members, their anti-reconnaissance methods are more sophisticated, so cracking criminal gangs is of great significance. [0003] In the prior art, determining criminal gang members is usually based on label propagation and clustering. The disadvantage of this method is: in the graph-based network, the label propagation is to find a structure with dense connections between graphs (or communities) and relatively sparse connections between different communities. This kind of structure mining idea does not match the gang mining, because in the network diagram of the public security system, there is often no direct relationship between gang members, and there may be few indirect relations...

AI Technical Summary

Problems solved by technology

The disadvantage of this method is: label propagation is in a graph-based network, looking for a structure with dense connections between graphs (or communities) and sparse connections between different communities

This kind of structure mining idea does not match the gang mining, because in the network diagram of the public security system, there is often no direct relationship between gang members, and there may be few indirect relationships, and there are also many common characteristics among gang members. Therefore, it is difficult to accurately identify gang members through label propagation and clustering

Images