Community Detection Based On Specific Topic And Analysis Of Information Propagation

With the rapid rise of Internet technology,online social network has become a popular means of communication,and is the rapid development.Now China's popular online social network mainly Sina Microblog,Tencent QQ,WeChat,renren network,etc........Online social network of people's actions every day to produce large amounts of data,and these data contain a wealth of information waiting for us to dig and analysis.Such as Sina Microblog users post Bowen,forward comments Bowen,concerned about other users,etc.These user behavior can be used to understand the user behavior patterns and user interests.By analyzing the community structure of social network,it is helpful to study the characteristics of social network topology,discover the mode of user aggregation and influencing factors,and promote the application of information retrieval,information recommendation,information dissemination control and public security incident control.Social networks can play a role in many areas,such as political aiding,marketing,to assist in the arrest of prisoners and other activities.In recent years,social networking has become an important issue in the social network."Amazon's recommendation system" is a very successful marketing case,Amazon's use of user browsing history and other user behavior information to recommend similar products to the user,so that sales soared.In this paper,after the study of the community discovery algorithm and information propagation model,the author realized the community detection framework based on user-published content in Sina Weibo,and the existing information propagation model can not reflect the true social network communication process,Thereby improving the independent cascade model,making it more in line with the real social network communication process.The main research work of this paper includes the following two aspects:(1)Realize the community detection framework based on microblog content under specific topic,and improve the shortcomings of the previous community detection algorithm based on network topology,but based on microblog content(2)The independent cascading model is improved to increase the community attribute on the node,so that the improved independent cascade model is more suitable for the real social network propagation process,and uses the information propagation maximization algorithm to verify the results.