An Efficient Feature Selection Technique to Enhance Spam Email Detection

Spam detection is a critical challenge in ensuring the security and reliability of e-mail systems. This work suggests an advanced Machine Learning (ML) model to mitigate Spam. The Spam ML detection model employs a novel feature selection technique that combines the strengths of the Firefly Optimization Algorithm (FOA) and Harris Hawks Optimization (HHO). The model uses Random Forest (RF), XGBoost, and Naïve Bayes (NB) for classification purposes. The experimental results demonstrate that the union of FOA and HHO selected features significantly enhance the classification performance of RF, XGBoost, and NB classifiers compared to using FOA or HHO individually. Notably, the RF classifier, leveraging the proposed feature selection method, attained the highest accuracy of 99.83%. This work underlines the efficacy of integrating FOA and HHO for feature selection in Spam detection and highlights the potential of RF as a robust classifier in this context.