作者: Tuan-Anh Hoang , Ee-Peng Lim
DOI: 10.1007/978-3-319-13734-6_1
关键词:
摘要: In this paper, we propose the Topical Communities and Personal Interest (TCPI) model for simultaneously modeling topics, topical communities, users’ interests in microblogging data. TCPI considers different communities while differentiating personal from those of learning dependence each user on affiliated to generate content. This makes existing models that either do not consider existence multiple or differentiate between community’s interests. Our experiments two Twitter datasets show can effectively mine representative topics community. We also demonstrate significantly outperforms other state-of-the-art topic tweet generation task.