Proceedings of the tenth ACM SIGKDD international conference on Knowledge discovery and data mining
Approximate object location and spam filtering on peer-to-peer systems
Proceedings of the ACM/IFIP/USENIX 2003 International Conference on Middleware
Hi-index | 0.00 |
Prior studies on collaborative spam filtering with near-duplicate similarity matching scheme mainly represent each email by a succinct abstraction derived from email content text. Since these abstractions of emails cannot fully catch the evolving nature of spams, we propose in this paper a novel email abstraction scheme, which considers email layout structure to represent emails. With the proposed abstraction, we design a near-duplicate matching scheme to efficiently match each incoming email with a huge spam database.