Identifying and Evaluating the Internet Opinion Leader Community Through k-clique Clustering

With the rapid development of the Internet technology, the Internet has become an important source of information for many acquiring the public sentiment. Opinion leaders play an important role in leading in the direction of the public opinion. In this paper, we drew the community components from the replies of every post in BBS according to the structure of the community in the network, and we came up with a method of extracting the opinion leader community (OLC) based on the hierarchical structure. In this way there were more overlapping appearances among members of the communities. Thus, the relationship between any two communities can be enhanced, which makes it easier to identify the OLC. Then, we analyzed the revolution of the OLC and put forward a time-dividing method of dividing the whole communities into different parts based on the characteristics of the post and the time period and gave the suitable measurement parameter to get the evolution result of the communities. Finally, experiments proved the efficiency of the OLC extracting method and the properties of the OLC revolution were summarized.