基于关联规则的垃圾短信分类器模型  

Spam Short Message Classifier Model Based on Association Rules

作　　者：张永军[1] 刘金岭[1] 高尚兵[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：《南通大学学报（自然科学版）》2014年第3期6-12,共7页Journal of Nantong University(Natural Science Edition)　

基　　金：国家级星火计划项目(2011GA690190);江苏省教育厅高校哲学社会研究项目(2012SJD87001);江苏省青蓝工程项目

摘　　要：针对垃圾短信分类问题,提出了一种通过挖掘垃圾短信关联规则来构造垃圾短信分类器,从而实现垃圾短信过滤的模型.该方法采用改进的FP-grow算法挖掘垃圾短信关联规则集,以关联规则集为基础构建垃圾短信分类器模型,在分类过程中考虑垃圾短信特征词权重和垃圾短信的变异行为进行分类.实验结果表明,该方法的分类精确率和非垃圾短信错分率优于其他常用垃圾短信分类方法.A classifier model based on association rules was proposed to filter spam short messages（SSM）. To mine association rules which were used to build the spam classifier model, an improved FP-grow algorithm was put forward. The factors of word weights and spam variation behaviors were considered to make SSM classifying more effective. The experimental results show that the proposed model is superior to other classification methods in the aspect of precision and misclassification rate of non-SSM.

关 键 词：关联规则 垃圾短信过滤 文本分类 垃圾短信变异 特征提取 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]