基于疫苗机制的垃圾邮件过滤模型

针对传统的垃圾邮件过滤方法不能有效阻止出现的新型垃圾邮件的问题,借鉴生物免疫系统中疫苗的概念和免疫记忆功能,提出了一种基于疫苗机制的垃圾邮件过滤模型SFM-V(spam filtering model based on vaccine mechanism).该模型详细描述了垃圾邮件检测器的演化和抗原提呈的过程,通过疫苗控制器实现疫苗提取和疫苗接种,新生成的免疫记忆细胞作为疫苗实现信息交互,共享抗体.并引入小生境免疫记忆与共享机制,增加检测器的多样性及稳定性,促进免疫记忆库及原始抗体库中优良个体的保存,有效地提取和记忆垃圾邮件的未知特征和变异特征.利用CCERT(china education and research network)的邮件样本集对该模型进行训练和检测,仿真实验结果表明该模型有效地提高了垃圾邮件的正确率、召回率等特性,降低了垃圾邮件的虚报率.

A improved spam filtering model based on vaccine mechanism

For the traditional spam filtering method could not effective prevent new spam problem ,referenced to the concept of the vaccine in biological immune system and immune memory function ,proposed a spam filtering model based on the mechanism of the vaccine . T he model described the evolution of spam detector and the process of antigen presentation , the extraction and vaccination vaccine was achieved by vaccine controller ,new generation of immune memory cells as a vaccine information interaction , sharing antibodies ;and the introduction of immune memory niche and the sharing mechanism ,increased the detector diversity and stability ,promoted the preservation of the excellent individual of the original antibody immune memory library and the library ,effectively extract and memory unknown and variation characteristics of the spam .Using CCERT mail sample set for training and testing the model ,simulation results show that the models effectively improve the accuracy , the recall rate and other characteristics of spam ,reducing the rate of false spam .