A COMPARATIVE STUDY ON SPAM EMAIL: DATA ANALYSIS BY VARIOUS CLASSIFICATION ALGORITHMS ALONG WITH JUSTIFICATION OF J48

Nowadays email becomes one among the fastest and most economical and effective media of communication. Hence as increase of email users dramatically increase of spam emails during the past few years. The data mining classification algorithms are classified into categorize this email as spam or non-spam. During this paper, we conducted experiment within the WEKA environment by using three algorithms namely Naive Bayes, J48, Support Vector Machine (SVM) on the spam email dataset and later the three algorithms were compared in terms of classification accuracy. The in-depth analysis of the study and descriptions of the three classification algorithms is presented consistent with our data simulation results the J48 classifier outstanding performs than Naive Bayes and SVM in terms of classification accuracy level performance.