A Novel Online Spam Filter Based on URLs and Maximum Entropy Model
Yang Li, Binxing Fang, Li Guo · 2006
Spam filtering is a great problem nowadays. The conventional spam filtering techniques still result in high false positives and false negatives. This paper proposes a novel online spam filter based on URLs and maximum entropy model. The filter identifies spam by classifying the e-mails with the pre-trained classifier based on the maximum entropy model and filters the spam online in terms of the characteristics of SMTP. Experimental results demonstrate it can significantly raise the filtering accuracy, effectively reduce false positives and can be applied to online processing environment by reducing the computational cost than the state-of-the-art techniques