Research on the mining of opinion community for social media based on sentiment analysis and regional distribution

Many researches have been carried out on network comments nowadays. In order to get more sentiment types, the paper put forward a opinion community discovery method based on sentiment of web comments and regional distribution, considering that Internet users from different regions have different perspectives on the same issue. First, the research clusters network comments into different opinion communities according to sentiment similarity. Then in the paper the Longest Sequential Sentiment Phrase (LSSP) is defined. The representative view of each pinion community is extracted by the mining algorithm of LSSP. At last, it studies regional distribution of opinion communities. Experimental results show that the method can be applied to analyze network comments. The research also shows that we can get more information from network comments than other related network public opinion analysis.