Community discovery method and device

Abstract

The invention discloses a community discovery method and device, belonging to the fields of computer technology and Internet technology. The method includes: dividing the network nodes in the social network into communities according to a preset community discovery algorithm, and obtaining n small communities and the label of each small community; n is an integer greater than or equal to 2, and the preset community discovery algorithm divides the community The accuracy of the community node is higher than the preset threshold; the label of the community node is updated according to the label propagation algorithm, the community node is a network node in n small communities, and the initial label of the community node is the label of the small community to which the community node belongs; Nodes are divided into a large community, and m large communities are obtained, m

Description

technical field [0001] The embodiments of the present invention relate to the fields of computer technology and Internet technology, and in particular to a community discovery method and device. Background technique [0002] Community discovery refers to dividing the social network into different communities, so that the user relationships within the same community are closely connected, while the relationship between communities and community fingers is sparsely connected. [0003] An existing community discovery method includes: the server divides the social network into communities according to the SCD (Scalable Community Detection, scalable community detection) algorithm, and obtains multiple communities. Wherein, the social network refers to at least one of relational networks such as Facebook, Weibo, campus network, and instant messaging application programs. [0004] Because the SCD algorithm is based on the triangles in the network, the internal links of the divided...

AI Technical Summary

Problems solved by technology

[0004] Since the SCD algorithm is based on the triangles in the network, the internal links of the divided communities are closely linked and have a certain degree of accuracy. However, because the community is too finely divided, it is not suitable for the needs of many scenarios.

For example, taking a social network as a campus network as an example, the divided community is usually as small as the department the user belongs to in the community, while for the campus network, users usually expect to be divided by student stages such as junior high school classmates, high school classmates, and university classmates , so it is obvious that the above division method does not meet the needs of users

Images