A Social Network Node Mining and Activation Method Based on Greedy Subgraph

Abstract

The invention provides a social network node mining method based on a greedy subgraph. Firstly, according to a node degree which is an important attribute, combined with the aggregation coefficient of a local topological structure, the influence potentials of nodes are estimated, the nodes are sorted according to high and low influence potentials and are added into a seed node candidate set, and through the overall evaluation sorting of a network, a node with a highest specificity threshold is selected to be added into the seed node candidate set. After completing the selection of the candidate set, through expressing a linear threshold model performance with improved influence as a greedy subgraph strategy, the real propagation simulation on the nodes in the set is carried out, a node with a largest incremental impact range is selected and added into a final node mining result set, the nodes in the candidate set are corrected dynamically when each step of propagation is completed, a candidate set correction process and a propagation simulation process are repeated until reaching the a node mining result set with an expected scale, and finally an ideal node mining effect is obtained.

Description

technical field [0001] The invention relates to a social network node mining method. Background technique [0002] Node mining methods in social networks are mainly divided into heuristic methods and greedy methods. The former mainly measures the importance of each node in the network based on the attributes of the social network nodes themselves or the topology of the network itself. For example, the degree centrality algorithm only considers the node’s neighbor topology when calculating the importance of nodes, although its calculation speed It is fast, but the accuracy is not good; another example is the proximity centrality algorithm and the betweenness centrality algorithm, because the calculation involves the entire network topology, so the algorithm efficiency is very low. The latter is to simulate the propagation of each node through the propagation model, and then calculate the importance of the node through the size of its propagation range. This type of algorithm...

AI Technical Summary

Problems solved by technology

The former mainly measures the importance of each node in the network based on the attributes of the social network nodes themselves or the topology of the network itself. For example, the degree centrality algorithm only considers the node’s neighbor topology when calculating the importance of nodes, although its calculation speed Fast, but the accuracy is not good; another example is the proximity centrality algorithm and betweenness centrality algorithm, because its calculation involves the entire network topology, so its algorithm efficiency is very low

The latter is to simulate the propagation of each node through the propagation model, and then calculate the importance of the node through the size of its propagation range. This type of algorithm is not applicable due to the combination of the propagation model for real propagation and the low efficiency of the algorithm. large social network

Images