Search Method And Organization Structure Discovery Of Network Spammers Based On Social Network

The network spammers is a cancer that affects the whole social situation and public opinion,and has great political and economic harm.Nowadays,the identification method of network spammers is mainly for single network spammer.It is a challenge for identification of large-scale network spammers to detect the network spammers from the individual perspective and ignore the group.There are two problems in the existing identification methods of network spammers:Firstly,the large number of network users makes it difficult to detect the network spammers in the whole network.At the same time,the number of social network user nodes is huge and the behavior of network spammers is dynamic,which leads to the real-time identification of searching the network spammers of the whole network will become a challenge.It's impractical to identify hundreds of millions of social users,just as you can't test everyone in epidemic prevention.In order to solve this problem,this paper proposes several algorithms of searching the network spammers based on social network.Search for other network spammers based on a few known network spammers It avoids the identification of the whole network node,which greatly improves the efficiency.BFS algorithm and DFS algorithm based on concern relation,as algorithm based on @ relation network spammers search algorithm,ES algorithm based on participation event search algorithm and MS algorithm based on three relationships are studied.The search algorithm extending from individual network to the whole network is studied.In the experiment,we first verify that BFS algorithm is faster than DFS algorithm;secondly,it verifies that several search algorithms have higher time efficiency and ensure accuracy than global search;third,it verifies that several search algorithms have faster search speed and higher accuracy than previous search algorithms.Secondly,a single network spammer can not easily affect social situation and public opinion.There are always a large number of organized groups of network spammers that affect social situation and public opinion.At the same time,it is difficult to find the organizational structure of network spammers.To solve this problem,this paper uses the social network analysis method to analyze the group structure of the group of network spammers,including point centrality analysis,intermediate centrality analysis,proximity centrality analysis and agglomerative subgroup analysis.Experimental results show that the method is effective.There is a central point in the network of network spammers,and there are many organizations of network spammers.