COMPARATIVE RESEARCH ON SHORT TEXT CATEGORIZATION

摘要：

Resent year,various kinds of short text data has been continually emerging in large numbers,such as article summary,e-mail,instant messages online and so on. Although the text categorization techniques in a number of areas have been gotten very good research and applications, for relatively short and different structures of the short data, the categorization research is relatively less.In fact, in order to use handily,it is very necessary to do the categorization research and application.In this paper,based on existing research,we analyzed and research some current text categorization algorithm, according to the experimental data,we also analyzed and compared that,in short text data categorization,SVM algorithm has good performance, and we suggested some improvements just for reference.

关键词： Short Tezt Categorization K-Nearest Neighbor (K-NN) Naive Bayes (VB) Support Vector Machine (SVM)

作者: Juan Chang Xueqin Lu

作者单位: Software college,Ningbo Dahongying University,Ningbo 315175,China Ningbo Dahongying University,Ningbo 315175,China

会议类型: 国际会议

会议名称: 2009 International Symposium on Computer Science and Technology(2009 中国宁波国际计算机科学与技术学术大会)

会议地点: 宁波

会议语种:英文

页码: 335-338

在线出版日期: 2009-11-20（万方平台首次上网日期，不代表论文的发表时间）