ABSTRACT
Email communication has become a fundamental part of modern personal and professional interactions. However, the proliferation of spam emails has given rise to numerous challenges, including information overload, reduced productivity, and heightened security concerns. Traditional rule-based methods for spam email detection have limitations in adapting to evolving spam tactics. This research presents a comprehensive study aimed at enhancing spam email detection systems through the integration of machine learning classifiers and Genetic Algorithmbased feature selection. The study explores the effectiveness of machine learning classifiers, including Naive Bayes, Support Vector Machines (SVMs), and K-Nearest Neighbors (KNN), in identifying spam emails. Additionally, it investigates the potential of Genetic Algorithms for optimizing feature selection, thereby improving classification accuracy. The research methodology encompasses data collection, preprocessing, model construction, and evaluation. A substantial dataset of spam and legitimate emails is utilized, and data preprocessing techniques such as tokenization, stop word removal, and stemming are employed. Models are constructed using SVM, K-Nearest Neighbors (KNN), and Naive Bayes classifiers, and their performance is evaluated using various metrics. Preliminary findings suggest that Genetic Algorithm-based feature selection enhances the accuracy of spam email detection systems. This research contributes to the field of email security by providing insights into novel approaches for addressing the challenges posed by spam emails. The potential implications extend to improving user experiences, safeguarding online communication, and enhancing the cybersecurity landscape.
