Loading...
Please wait, while we are loading the content...
Similar Documents
Community Discovery in Twitter Based on User Interests
| Content Provider | CiteSeerX |
|---|---|
| Author | Zhang, Yang Wu, Yao Yang, Qing |
| Abstract | Twitter has recently emerged as a popular social microblogging service. There are over 100 million users in Twitter nowadays, little is known yet about Twitter at user level. In this paper, we investigate the problem of identifying communities in Twitter based on users ’ interests. To address this problem, we first compute user similarity leveraging both textual contents and social structure, according with Twitter’s role, not only a news media but also a social network. These features include tweet text, URLs, hashtags, following relationship and retweeting relationship, and all of them are closely correlated with users ’ interests. Then we use user similarity as well as classical clustering algorithms to discover communities. To assess effectiveness of our method, we propose the metrics in Twitter “average number of mutual following links per user in per community”. Experimental results show that our method can successfully discover communities in Twitter, and gives a much better performance than random selection. From a side view, our experiment also shows that users in our dataset of Twitter can be approximately categorized into 400 communities. |
| File Format | |
| Journal | Journal of Computational Information Systems |
| Language | English |
| Access Restriction | Open |
| Subject Keyword | User Interest Community Discovery User Similarity User Level Popular Social Microblogging Service Twitter Average Number Random Selection News Medium Textual Content Social Network Experimental Result Twitter Nowadays Tweet Text Retweeting Relationship Twitter Role Side View Social Structure |
| Content Type | Text |
| Resource Type | Article |
