A Phishing Email Detection Model Based on Horse Herd Optimization and Random Forest Algorithms
P. Hemannth, Mukesh Chinta, S. Sarat Satya, P. Sri Aneelaja Devasena · 2024
The ever-growing threat of phishing emails presents a considerable hurdle to the security of individuals and organizations. In response, researchers have explored innovative approaches to enhance email security. This paper proposes an innovation approach for identifying phishing emails by combining the Horse Herd Optimization algorithm with the Random Forest algorithm. Horse Herd Optimization, inspired by the collective behavior of horse herds, is employed to optimize the feature selection process, aiming to identify the most relevant attributes for discriminating between legitimate and phishing emails. Subsequently, the selected features are fed into a Random Forest classifier, a powerful ensemble learning technique, to classify emails into their respective categories. The experimental results, obtained using a comprehensive dataset of real-world phishing emails demonstrate the effectiveness of the proposed approach. The hybridization of HHO and Random Forest yields a robust and accurate phishing detection system, outperforming traditional methods in terms of both accuracy and efficiency. The integration of nature-inspired optimization and machine learning techniques showcases the potential of interdisciplinary solutions in addressing complex cybersecurity challenges. This research contributes to the progression of email security and lays the groundwork for crafting more sophisticated and adaptive anti-phishing systems.