Email Spam Classification Based on Logistics Regression

Email is a common communication tool used by both individuals and organizations. It involves a variety of interactions, including file sharing. In addition to the advantages it provides, there is the uninvited email sharing. This unsolicited email is referred to as "spam." Malicious content like viruses, phishing scams, and unsolicited ads can be found in spam. It is well known that communication security is crucial. In order to filter email systems for malicious tools or software, it is essential to classify them based on a variety of criteria. In these kinds of classification studies, machine learning algorithms work well. The objective of this research is to solve the problem at hand and compare the logistic regression, random forest, naive Bayes decision tree, and support vector machine (SVM) algorithms. The effects of various methods and approaches on the issue were thoroughly examined. A comparison of the various performance outcomes using the various approaches is provided. With an accuracy of 98%, logistic regression was the most accurate, followed by random forest with 97%.