Spam Email Detection on Data Mining: A Review

Abstract

As we know email is an effective tool for communication and it is the fastest way to send information from one place to another and it saves time and also cost. But the email is affected by attacks which include spam mails. Spam is unwanted email or it is bulk data that is flooding the internet with many duplication of similar message, in an attempt to force the email on people who would not otherwise choose to receive it. To address the growing of spam email on the internet the interest of spam filtering also grow accordingly. In this paper we review various spam detection technics. We are use the technics with feature selection algorithm and without feature selection algorithm and apply all the classifier of data mining tool. In this study we analyze the classifier algorithm using two different data mining tools those are WEKA and TANAGRA. Data mining is the discovery of knowledge from the large database and it is the technique of finding out new patterns in a huge data sets. Both data mining tool use different classification algorithms like K-Nearest Neighbor (K-NN), Naïve Bayes (NB) and others. Then finally, the best classifier for email spam is identified based on the accuracy of the algorithm on each data mining tools.

Keywords: Classifier, Feature selection, Spam E-mail.

DOI: 10.7176/JIEA/9-2-01

Publication date: April 30^th 2019