Cluster type identification method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a cluster type identification method, a device, electronic equipment and a storage medium. In the method, obtaining a to-be-tested feature of the to-be-testedcluster, wherein the to-be-tested feature at least comprises an importance degree parameter corresponding to each preset behavior parameter of each member of the to-be-tested cluster, which indicatesthat the to-be-tested feature can represent the behavior feature between the members in the to-be-tested cluster and the importance degree of each member on each preset behavior parameter in the to-be-tested cluster; as the cluster type prediction model is obtained by training the neural network, the cluster type prediction model can obtain the cluster structure of the to-be-tested cluster based on the to-be-tested feature in a multi-angle and multi-dimensional manner; and in the process of training the cluster type prediction model, the minimum length of the real vector and the difference vector is used as a training target, so that the cluster type prediction model can accurately predict the prediction probability that the to-be-tested cluster belongs to various known clusters.

Description

technical field [0001] The present application relates to the field of communication technology, and more specifically relates to a cluster type identification method, device, electronic equipment and storage medium. Background technique [0002] The current network is composed of several groups, and a group is a combination of a certain type of members with the same characteristics. For example, the group may be a QQ group or a WeChat group. The association between members in each group determines whether each group belongs to the same community, for example, the ratio of the number of overlapping members corresponding to each group in the same community to the total number of members of the corresponding group is greater than or equal to the preset threshold , such as a community includes two QQ groups. [0003] The group structure of a group can reflect the behavior characteristics of each member in a group. The community structure of a community can reflect the behavi...

AI Technical Summary

Problems solved by technology

[0005] In view of this, the embodiment of the present invention provides a cluster type identification method, device, electronic equipment and storage medium to overcome the problem of low efficiency in the prior art

Images