Research on Discovery of BBS Topic Based on LSA
Geng Huan-tong · Computer Knowledge and Technology · 2008
To obtain BBS topic effectively has been an important research field of web information processing. By analyzing the shortage of existed topic discovery algorithms, a new discovery method is proposed in the paper. The method is based on latent semantic analysis, by calculating the similarity between comments, clustering the topics with the factors of time and space. Finally the visual and interactive sys-tem is implemented, which can discovery topic drift and new topic timely and effectively.