%0 Journal Article %A QIAN Yunyun %A YANG Wenzhong %A YAO Miao %A LI Hailei %A CHAI Yachuang %T Topic Community Discovery Model Incorporating Topic Similarity Weight %D 2021 %R 10.3778/j.issn.1002-8331.1911-0452 %J Computer Engineering and Applications %P 107-114 %V 57 %N 5 %X

Social network structure is complex, and the topic community is one of the important ways for personalized recommendation and business promotion. However, the existing methods of topic community mining are either only based on link relationship and text information mining topic community, or mining the topic based on the divided community, the interaction between topic and the community is ignored, which results in the low similarity of the topic within the community. Therefore, it proposes a new community topic computing method, then establishes a topic community discovery model (TSWTCD) that integrates topic similarity weights. The topic is extracted by text information, the topic similarity between nodes is calculated as link weight, and link weight is taken as module parameter to divide the community. Finally, the community topic is obtained according to the proposed new community topic calculation method. Experimental results based on real data set show that the TSWTCD model improves the quality of mining topic communities.

%U http://cea.ceaj.org/EN/10.3778/j.issn.1002-8331.1911-0452