Improving email spam classification performance using machine learning techniques

Abstract

newline Over the past few decades, communication has become easier newlinedue to the rapid development of technology. Although several modes exist newlinefor communication, in this era of the Internet, Electronic mail, or Email, turns newlineout to be very popular because of its effectiveness, affordability, and ease of newlineuse for personal communication as well as business purposes, as well as newlinesharing important information in the form of text, images, documents, etc. newlinewith others. This proficiency leads to email being exposed to numerous newlineattacks, including spam. The emails that are useful to the user are referred to newlineas ham or legitimate emails, whereas the unwanted emails are known as spam newlineemails. newlineAt present, spam email is a major source of concern for email newlineusers, where unbidden messages, used for business purposes, are directed newlineextensively to several mailing lists, entities, or newsgroups. These spam newlineemails are used for the purpose of advertising products, collecting personal newlineinformation, sending destructive contents in the form of executable files to newlineoutbreak user systems, or providing a link to a malicious website to steal newlineconfidential data, such as bank accounts, passwords, etc., leading to a newlinereduction in efficiency, security threats, consuming server storage space, and newlineunessential consumption of network bandwidth. Spammers collect email newlineaddresses from chat rooms, websites, customer lists, newsgroups, and viruses newlinethat harvest user addresses. Spam emails are sent out in bulk quantities every newlineday, and these spam emails often have very similar characteristics, allowing newlinethem to be detected using various machine learning and deep learning newlinealgorithms.