Community division method and device

Abstract

The invention discloses a community division method, which comprises the following steps that: receiving the information of multiple uses to be divided, and establishing a topological structure diagram according to the information of multiple uses to be divided; calculating the degree of each node in the topological structure diagram, taking the node with a highest degree in the topological structure diagram as a seed node, and obtaining the neighbor node of the seed node; according to the neighborhood attribute of each neighbor node, calculating the input vigor proportion of the seed node for each neighbor node; according to the input vigor proportion of the seed node for each neighbor node, through a grid constraint algorithm of a structural hole, obtaining the constraint factor of the seed node for each neighbor node; and dividing the neighbor node of which the constraint factor is greater than a preset relationship threshold value and the seed node into the same association, and deleting the neighbor node of which the constraint factor is greater than the preset relationship threshold value and the seed node from the topological structure diagram to improve community division quality.

Description

technical field [0001] The invention relates to the field of computers, in particular to a community division method and device. Background technique [0002] Community structure is a common property of social networks, which satisfies the characteristics that the internal nodes of the same community are relatively closely connected and the nodes of different communities are relatively sparsely connected. The essence of community division is to divide the social network into a smaller community, the members within the community are relatively close, and the membership between communities is relatively sparse, which is very similar to the essence of cluster analysis. [0003] The typical representative of the agglomeration algorithm of the existing community division research algorithm is the Newman fast algorithm. Its idea is to start from an empty network and gradually add similar edges. In order to remove all the edges in the network, each node of the network is regarded ...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the embodiments of the present invention is to provide a method and device for community division, which can effectively solve the problem that the existing community division methods ignore the neighborhood information of neighbors, and achieve a significant increase in the efficiency and accuracy of community division under specific data.

Images