Spam Email Detection and Categorization using Deep Learning Techniques

Abstract

Email has become one of the most widely used forms of communication. Email spam refers to unsolicited messages sent in large volumes. While some spam emails may contain useful information, most are unwanted and can lead to online fraud. Therefore, filtering spam emails from legitimate ones is essential. Effective categorization of spam emails enhances cybersecurity, reduces inbox clutter, and enables better filtering of malicious, promotional, and irrelevant content. In this paper, a deep learning based approach is proposed to categorise the spam emails into promotion, marketing, news, security, and others. The proposed approach uses Support Vector machine and Word2Vec with the Bi-directional Long Short Term Memory Network techniques. Experimental results indicate that the Support Vector machine technique gives better performance than the other models with an accuracy of 99%. Further, the proposed system offers a robust solution for spam email detection and categorization, which can be effectively implemented in email platforms to enhance security, improve user experience, and reduce the burden of unwanted content.