融合hownet和btm模型的短文本聚类方法研究.pdfVIP

M ay COM PU T ER ENGINEERING A ND DESIGN Vol No HowNet BTM How Net BT M CSPA Biterm T P A doi issn j Shorttex t clustering method combining How Net w ith BT M model Y A N G Xiaolan Y A NG Wei QIA N Cheng Z H U Fuxi School of Information and Engineering Wuchang U niversity of T echnology Wuhan China Computer School Wuhan U niversity Wuhan China Abstract Using conventional techniques in the short text clustering the similarity calculation accuracy is poor and the clustering result is unstable A similarity calculation method based on How Net semantics thesaurus and BT M topic model w as proposed Both of them w ere linearly combined and short text similarity w as comprehensively studied Evaluation based on the clustering quality and clustering degree of difference w as established it w as used to evaluate advantages and disadvantages Better quality results w ere filtered out clus tering integration w as realized using CSPA fusion algorithm Experimental results show that the proposed method improves the accuracy of the calculation of shorttext similarity and improves the stability of the fusion results Key words shorttex t How Net BT M clustering fusion 0 How Net BT M