Content Based Spam Classifying Algorithms in Email
Vanessa Sunjaya, Stefan Senjaya, Joko Utama, Henry Lucky, Derwin Suhartono · 2022
The era of globalization has brought upon an era which we have never seen before. From a year, a month, a week, to a fraction of a second to send a message across the globe. The technological breakthrough of emails has simplified information deliveries and exchanges. Though an innovation means a new threat has also emerged in the form of spam emails. Spam emails are meant to target the vulnerability of the system in which anyone could send an email to someone if they knew their email address. These spam emails contain unwanted dangers that could prey on an unknowing victim. To combat this threat, spam classifying algorithms are developed, a method that could distinguish a spam email when it passes through. This paper will do a review on three content-based e-mail spam filtering, which are Support Vector Machines, Random Forest, and Multimodal Naïve Bayes. Comparing the effectiveness and usability of all three methods towards their capabilities and disadvantages were done. The experiment results in Support Vector Machines as the most efficient and the rest falling behind albeit by a fraction. Proving that even the smallest details matter in classifying algorithms.