Enhanced Naive Bayes Classifier for Email Spam Filtering Using the Product Rule of Logarithm

Abstract

The Naive Bayes algorithm is a widely used classification algorithm with applications in various domains. The prevalence of email spam poses a significant challenge to effective email communication every day. The study proposed an enhanced Naïve Bayes classifier for email spam filtering, utilizing the product rule of logarithm to overcome the issue of probability calculations. The methodology involves data preprocessing, application of log probability calculation, and training model classier using the enhanced algorithm. Provided a testing dataset that was used to evaluate results and demonstrate improved classification accuracy in the implemented filtering model. The proposed enhancement contributes to the robustness of the Naive Bayes algorithm in various classification tasks.