Classification of Spam and Ham Emails with Machine Learning Techniques for Cyber Security
Subrata Datta, Shaon Bandyopadhyay, Bappaditya Mondal · 2023
People often receive spam emails that may cause of cyber threats. Hence, classification of spam and ham emails is necessary for cyber security. Lack of classification accuracy is one of the common problems in this context. This paper proposes a new method of spam and ham classification based on machine learning (ML) techniques. The proposed method considers TF-IDF technique for feature selection. In this concern, flexible threshold based pruning in TF-IDF technique is introduced. In addition, the proposed method takes account of six popular ML techniques to classify the emails. Experimental results reveal that the flexible threshold helps to improve the classification accuracy of the ML techniques. The proposed method declares Support Vector Machine (SVM) as the best classifier with accuracy of 96.22%. Thus, the proposed method helps the users in identification of spam emails to avoid cyber threats.