A Machine Learning Approach for Identification of Spam Content in Email

Abstract

The internet has become the most important component of our lives, and it is used for everything. One of the most important applications of this is the exchange of information from one person to another. The increase in internet usage has resulted in an exponential surge of spam in the internet world. E-mail is the most often used online communication tool. The emails contain some unsolicited messages labelled as spam, which causes problems for consumers and necessitates the usage of dependable anti-spam filters. Many methods for detecting spam in email have been investigated. Spam consists of text and graphics that can affect the system. Spam senders grossly abuse email by broadcasting unsolicited facts. As a result, spam is one of the most common issues that an internet user must deal with. This paper proposes two classification methods for spam email detection: k-nearest neighbor's algorithm (KNN) and support-vector machines (SVM). During this process, the dataset is divided into many sets and fed into each algorithm. The findings of three studies are compared in terms of precision, recall, accuracy, f-measure, true negative rate, false positive rate, and false negative rate.