Analysis of Spam Detection Techniques using Machine Learning

Emails are used in practically all fields, including business and education. With the rapid growth in internet users, email spam has grown to be a serious concern these days. They are being used by people for fraud, phishing, and other unethical and illegal activities, using spam emails to send malicious links that could compromise both your and our systems. It is significantly easier for spammers to create a phoney profile and email account; in their spam emails, they pose as real persons, and they prey on those who are unaware of these frauds. User security is at risk when a large volume of unsolicited emails are sent. Spammers create a great deal of internet vulnerability despite several security measures. Ham and spam are the two subcategories of emails. Email spam, sometimes referred to as junk mail or unsolicited emails, is a kind of email that can be used to hurt users by robbing them of vital information and squandering their time and computer resources. Spam emails are becoming more and more prevalent every day. These days, email and IoT service providers face serious and massive challenges with spam identification and filtering. Of all the methods created to identify and stop spam, email filtering is one of the most important and well-known. Various machine learning and deep learning methodologies have been employed for this objective. The best algorithm is chosen for email spam detection with the best precision and accuracy in this study, which will also cover machine learning algorithms and apply all of these algorithms to our data sets. The machine learning algorithms used to design a perfect machine learning model for spam detection in emails include: Naïve bayes, SVM , Decision tree, KNN , Logistic Regression are analyzed in this work.