中文摘要：

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．

英文摘要：

Considering concept drift of the spare, this paper analyzes the disadvantages of the spare filtering algorithms on use and pmposes a spare filtering algorithm based on CBR（Case-base Reasoning）. After identifying features of spare, 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 spain filtering algorithm have been presented and the results show that the algorithm achieves better performance of changing spain filtering compared with the traditional Bayes.