Analyzing the Effectiveness of N-gram Technique Based Feature Set in a Naive Bayesian Spam Filter
Nikhil Mathew, V. Ramani Bai · 2016
The advent of Social Medias, Email services and other internet facilities are found helpful for a wide range of users. But some of them are interested in finding loop holes in such web based services to hinder the normal activities of common users. In this, spam Emails are one of the most disturbing activity in social network. In this context there is a need for efficient spam filters and most of the Email service providers dispense their own filtering mechanisms to deal with spam mails. For building efficient spam filters a strong feature set is required in order to train these filters. In this survey, a feature set created using N-grams technique was used as a training set for Naive Bayesian classifier and analyzed its classification efficiency for different ranges of grams. Based on the study a hybrid email spam filter model based on sentiment analysis was proposed as a future work.