Machine Learning Approaches for an Automatic Email Spam Detection

Abstract

With the rapid growth of internet users, spam emails have become a major problem. Spammers can easily create fake profiles and email accounts by pretending to be genuine people in the sent emails. The spammers target people who are unaware of such scams. In today’s environment, email is a simple, quick, and cost-effective way to communicate but has various security threats which are necessary to identify to maintain security. This situation necessitates having an inbuilt spam filtering system to use email effectively without being worried about losing personal details. The goal of this work is to discover and predict spam emails early by using various classifiers. Machine learning methods provide the most accurate spam classification. This article contributes towards the development of a spam detection model by using multiple classification methods to tackle spam email challenges and helps in the technological progress in privacy & security. This model employs classification technologies such as naive bayes, K*, J48, and random forest. Conclusively, when the random forest model has been used as a prediction classifier, the output of this model has shown the highest accuracy of 95.48%.