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.