An in-network spam filtering method using nearest neighbor search of edit distance-猫眼课题宝

一种利用编辑距离最近邻搜索的网内垃圾邮件过滤方法

DOI：

发表时间：

2011

期刊：

影响因子：

通讯作者：

中科院分区：

文献类型：

作者：

Tsukajima Yosuke;Nakao Akihiro

大量的垃圾邮件已经成为这种通信手段的严重威胁，因为根据现有的研究[1]，垃圾邮件占整个电子邮件消息的80%之多。大量的垃圾邮件从僵尸网络发送，广泛分布在互联网上。虽然传统的垃圾邮件过滤技术往往是在互联网的边缘实现的。在互联网的十字路口检测垃圾邮件是有意义的，而不仅仅是在边缘，因为大量的垃圾邮件是从僵尸网络发送的。在本文中，我们提出了一种新的网络中的垃圾邮件过滤技术，检测重复的内容的数据包流在网络的核心使用编辑距离。使用真实的数据包跟踪对所提出的方法进行的评估表明，我们可以成功检测垃圾邮件，召回率为88%，准确率为86%。我们还进行了可行性研究，我们的方法，特别是从角度来看，我们是否可以执行我们的方法在实时高速路由器和

A huge amount of spam has become a serious threat to this communication means, since according to the existing research [1], spam accounts for as much as 80% of the entire Email messages. And a large amount of spam sent from bot-nets that are distributed widely across the Internet. Although the conventional spam filtering techniques are often implemented at the edge of the Internet. It makes sense to detect spam at the cross-road of the Internet, not only at the edges, because a large amount of spam sent from bot-nets. In this paper, we propose a novel in-network spam filtering technique that detects duplicate content of the packets flowing at the core of the network using edit distance. Our evaluation of the proposed method using the real packet trace shows that we can successfully detect spam with recall of 88% and precision of 86%. We also conduct feasibility study of our method, especially from the perspective as to whether we can perform our method in real-time at high speed routers and

An in-network spam filtering method using nearest neighbor search of edit distance

一种利用编辑距离最近邻搜索的网内垃圾邮件过滤方法

文献摘要

被引文献

相似文献