Study on Short Text Classification with Integrated Algorithm

摘要：

　　With the rapid growth of the number of short text, how to effectively realize the automatic classification of short text is needed to be solved in the information domain.According to the characteristics of short text, this paper proposes Bagging_NB & Bagging_BSJ, which are two classification algorithms based on the improvement of current integrated classifiers.Traditional classifier NB, SVM, J48 are used as the basis classifiers to train the classification models.Compared with several individual classifiers, our methods have excellent results in a variety of classification evaluation indexes.

关键词： text classification short text integrated algorithm

作者: Dexin Zhao Nana Du Liangliang Qin

作者单位: Tianjin Key Laboratory of Intelligent Computing and Novel Software Technology, Tianjin University of Technology, Tianjin 300384, China

会议类型: 国际会议

会议名称: The 13th Web Information Systems and Applications Conference(第十三届全国web信息系统及其应用学术会议)(WISA2016)、The 1st Symposium on Big Data Processing and Analysis)( BDPA 2016)第一届全国大数据处理与分析学术研讨会、The 1st Workshop on Information System Security)(ISS2016)(第一届全国信息系统安全研讨会

会议地点: 武汉

会议语种:英文

页码: 121-124

在线出版日期: 2016-09-23（万方平台首次上网日期，不代表论文的发表时间）