Advanced Email Spam Detection: A Machine Learning Solution

Email categorization is crucial in business and academia, filtering spam emails that risk phishing, fraud, and theft. This study employs a hybrid approach, considering group and individual strengths and weaknesses in email classification. It prioritizes identifying spam, particularly phishing, using data mining and the UCI spam base dataset. The primary objective is to enhance automated learning for accurate detection and filtering, differentiating legitimate messages from spam. Radial Basis Function Neural Networks (RBFNN) are central to this work, a key component of artificial neural networks (ANNs) for data classification. The paper encompasses various algorithms: Support Vector Machine (SVM), Neural Network, Naive Bayes, K-Nearest Neighbor (KNN), Decision Tree, Random Forest, Harris Hawk’s Optimizer (HHO), and Extreme Learning Machines (ELM). This research advances email classification, bolstering digital communication's security and efficiency.