Analysis of the user behavior and opinion classification based on the BBS

BBS is an electronic information center as well as an emerging media. BBS forums and other online media has become an important channel of expressing public opinions addition to the traditional mass media, the expression of popular channels. In this paper, we made the research work focused on BBS user behavior and opinions by using of the mining techniques, the result of which are very practical significance for government making decisions, the network purification and prosperity, and building a harmonious society. To study and mine the frequent behavior of users, this paper mines and analyzes the regular pattern of users visiting BBS Also, this paper mines the posts on BBS by ARC-BC text classification algorithm, which is based on associated rules, and divides user opinions into three categories, support, oppose and neutral. In this paper, a special generalization process and a frequent-set based analysis method are proposed according to the particular data type of user visit records. The experimental results show that certain users and the collective performance have the obvious similarity or difference, the algorithm has a good performance on the BBS text classification problem. Such data mining method can cluster users with similar behavior patterns as well. This paper shows the frequent-set based method is very effective for mining BBS user access patterns.