An Evaluation on the efficiency of E-Mail Spam Detection Using Naive Bayes Classifier

Abstract

Nowadays, electronic mail is ubiquitous, being used everywhere from the business sector to the classroom. Emails can be broken down into two distinct categories: ham and spam. Email spam, also known as junk email or unwanted email, is a form of email that can be used to harm any user by wasting his or her time, draining system resources, and stealing sensitive data. Every day, the proportion of spam emails increases dramatically. Today's email and IoT service providers face a large and formidable task in detecting and filtering spam. One of the most prominent and widely-known approaches of detecting and avoiding spam is email filtration. It's also one of the most discussed tactics out there. Several machine learning and deep learning techniques, including Naive Bayes, decision trees, neural networks, and random forests, have been employed to reach this objective. After completing a survey of the available machine learning approaches, this article groups them into the most acceptable categories for usage in spam screening on email and IOT platforms. Accuracy, precision, memory requirements, and other metrics are used to thoroughly assess the methodologies. In the final section, we examine both the overall takeaways and directions future studies could go in.