Edge clustering coefficient-based social network group division method

Abstract

The invention relates to an edge clustering coefficient-based social network group division method. Specifically, the method comprises the following steps: reading the social network data; constructing a social network planning which takes social network users as nodes and takes user relationships as edges; randomly endowing each user with a unique label value; updating the labels of the user nodes by adopting an edge clustering coefficient-based label propagation algorithm; and after several iteration, owning, by the tightly connected nodes, same specific label value. By adopting the social network group division method, the user groups are divided through improving the label propagation algorithm according to the edge clustering coefficient attribute of the user relationship graph, and the division result has preferable application value for the monitoring of network public opinions and the searching of commercial customers.

Description

technical field [0001] The invention relates to the technical field of social networks, in particular to a social network group division method based on edge clustering coefficients. Background technique [0002] How to mine useful information from social networks has become a research hotspot in complex networks. It is of great significance both in theory and in social practical value. Network communities are usually composed of network nodes with similar functions or properties. By mining the community structure in the network, users can quickly and accurately find relevant users who have internal connections with themselves, such as users with the same or similar interests. Network public opinion monitoring, commercial user mining and other fields have good application value. [0003] So far, people have proposed many community discovery methods. In 2002, Girvan and Newman published a paper on PNAS to study the community structure in social networks and biological netwo...

AI Technical Summary

Problems solved by technology

Typical representative algorithms include Newman fast algorithm, GN algorithm, etc. The disadvantage is that the algorithm has high complexity and cannot define when to stop

[0004] It can be seen that the above classic algorithms have many limitations, the division results are not ideal, and the complexity is high, it is difficult to meet the requirements of large-scale real network community discovery

In 2007, Raghavan et al. proposed the label propagation algorithm, which effectively solved the problem of high complexity and inability to converge.

[0005] However, although the label propagation algorithm is simple and efficient, the randomness of the label propagation in the algorithm leads to poor accuracy of the algorithm, unstable division results, strong randomness, and robustness to be improved

In summary, there is a lot of room for improvement in the accuracy and time complexity of existing community discovery methods.

Images