An efficient LPP-based spam filtering algorithm

Abstract

With the fast expansion of the Internet globally in the last decade, the spam e-mail has become a main problem of the email service for Internet service providers, corporate and private users. To efficiently solve the spam filtering problems, a spam mail filtering method based on locality pursuit projection (LPP) and nearest feature line (NFL) classifier is proposed in this paper. Experimental results show that the proposed algorithm achieves much better performance than other traditional spam filtering methods.