一种改进的高效贝叶斯短信文本分类器  

An Improved Efficient Bayesian Short Message Text Classifier

作　　者：张永军[1] 刘金岭[1] 

机构地区：[1]淮阴工学院计算机工程学院,江苏淮安223003

出　　处：《南京师范大学学报（工程技术版）》2014年第3期70-74,共5页Journal of Nanjing Normal University(Engineering and Technology Edition)

基　　金：国家级星火计划项目;农村民生建设信息反馈平台建设项目(2011GA690190)

摘　　要：针对短信分类问题,提出了分类能量空间的概念,将特征词转换为分类能量空间上的一个能量元,以此为基础计算短信的能量特征向量.通过计算短信能量特征向量的领域密度,结合贝叶斯公式输出了短信在不同分类的分类概率.在分类过程中,还对分类概率差别较小的短信采用支持向量机进行了二次分类以提高分类效果.实验结果表明,该分类器模型具有良好的分类效果.A Bayesian classifier model is proposed to classify short message according to its content. The concept of category energy space is introduced and the word feature is converted to an energy unit in category energy space. Then the short message is represented as an energy vector based on its words. To obtain each category’s probability, the energy vector density is calculated and brought in Bayesian probability formula. When the category probabilities are not very different,a SVM model is used to reclassify the short message. The experimental results shows that the proposed model is superior to other classification methods in the classification result.

关 键 词：短信 文本分类 贝叶斯 支持向量机 分类能量空间 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]