A Comparative Impact Study of Attribute Selection Techniques on Naïve Bayes Spam Filters

The main problem of the Internet e-mail service is the massive spam message delivery. Everyday, hundreds of unwanted and unhelpful messages are received by Internet users flooding their mailboxes. Fortunately, nowadays there are different kinds of filters able to identify and automatically delete most of these messages. In order to reduce the problem dimensionality only representative attributes are selected from each e-mail using feature selection techniques. This work presents a comparison among five well-known feature selection strategies when they are applied in conjunction with four different types of Naïve Bayes classifiers. The results obtained from the experiments carried out show the relevance of choosing an appropriate feature selection technique in order to obtain accurate results.