EMAIL CLASSIFICATIONS FOR SPAM MAIL DETECTION BY COMPARING THREE DIFFERENT ALGORITHMS USING WEKA

Email is the system for sending messages from one individual to another via telecommunications links between computers or terminals using dedicated software applications. Nowadays, Email is used most common and effective mode of communication way to communicate in personal, individual and professional level. As increase of email users there will be increase of spam emails from the past few years. This paper explore how email data was classified using three different classifiers (Naive Bayes classifier ,Support Vector Classifier,J48 Classifier) for detecting spam using WEKA. This experiment was performed based on dataset to find spam in different parameters like finding Accuracy,Recall,Precision,Fmeasures and False Position Rate etc. The final classification result should be ‘1' if it is finally spam present , otherwise, it should be ‘0' for no spam. Finally this paper shows that J48 classifier is best and efficient algorithm for detection of spam emails for dataset that classified as binary tree among other algorithms.