Hybrid Machine Learning for Feature-Based Spam Detection

Researchers have been paying significant attention to the increasing need for and prevalence of unwanted spam on networks. Multiple studies have extensively examined the fundamental origins of spam in networks. Previous research has shown that spam-related problems are one of the main dangers to network systems, affecting both ongoing operations and past data. To tackle and reduce the increase in spam, researchers have put forward different methodologies. Although certain techniques have received approval from experts, every method has its own set of constraints. Furthermore, the persistent issue of inadequate precision in identifying spam poses a significant obstacle for numerous current predictive models. Therefore, the task of finding a reliable prediction method continues to be a challenging and extensively studied field, attracting significant attention from researchers in recent decades. This paper introduces a machine learning approach for detecting spam in emails. The proposed model demonstrates improved accuracy compared to previous studies.