基于事例推理的中文垃圾邮件过滤

分析现有垃圾邮件过滤分类算法的不足,根据垃圾邮件的概念漂移特性,提出了一种基于CBR的垃圾邮件过滤算法.针对中文垃圾邮件特点提取特征,设计基于CRN网络的实例检索算法,该算法增加了预计算阶段,从而提高检索速度.实验结果表明,与传统贝叶斯算法相比,该算法对于动态变化的中文垃圾邮件数据集有更好的过滤效果.

Chinese Spam Filtering Algorithms Based on Case Base Reasoning

Considering concept drift of the spam, this paper analyzes the disadvantages of the spam filtering algorithms on use and proposes a spam filtering algorithm based on CBR(Case-base Reasoning). After identifying features of spam, this paper designs an alternative similarity retrieval algorithm based on Case Retrieval Nets (CRNs), which has pre - computation phrase for speeding up retrieval . Experiments on the spam filtering algorithm have been presented and the results show that the algorithm achieves better performance of changing spam filtering compared with the traditional Bayes.