Enhancing E mail Spam Detection Efficiency Using Naïve Bayes Classifier

Abstract

In modern times, many individuals rely on easily accessible e mail or messages from random people. Because every individual may leave an e mail or message, spammers have a fantastic chance to construct spam messages relating to our different interests. Spam overflows mailboxes with nonsensical e mails all the time. It drastically lowers our internet speed. It contains crucial information, including the specifics of our contact list. Finding these spammers and the spam content can be difficult, and the matter is now under study. An operation to distribute messages in mass via mail is known as "e mail spam". E mail spam detection is vital for maintaining secure and efficient communication. This research paper investigates the application of the Naive Bayes classifier for e mail spam detection. It compares its performance with other machine learning algorithms, including Support Vector Machines (SVM), KNN and Decision Trees. Using the SMS spam collection dataset, models were trained and evaluated based on accuracy and precision. The Naive Bayes classifier demonstrated notable effectiveness, achieving an accuracy of 98.5% and a precision of 97.2%. Compared to SVM, KNN, Decision Trees and other machine learning algorithms, the Naive Bayes classifier showed competitive results, highlighting its simplicity and efficiency. This study underscores the potential of the Naive Bayes classifier as a reliable tool for e mail spam detection. It provides insights into the comparative performance of different algorithms in this context.