Advanced Intrusion Detection System Through Hybrid Integration of Random Forest and Deep Learning with Artificial Neural Networks
B. Prameela Rani, Ch.Sowndarya Lahari, J.Ananda Lavanya, Annemneedi Lakshmanarao, Gopala Varma Kosuri · 2024
The growing complexity of cybersecurity threats calls for advanced Intrusion Detection Systems (IDS) that surpass the capabilities of traditional methods. This paper introduces a hybrid approach combining ML and DL techniques to enhance IDS accuracy. Initially, we applied several base ML models, including AdaBoost, Random Forest (RF), Logistic Regression, and Gradient Boosting, to two datasets sourced from Kaggle. Among these, the Random Forest model achieved the highest accuracy. Building on this, we developed a hybrid ML+DL model by combining the best-performing RF model with an Artificial Neural Network (ANN). This hybrid approach leverages the feature extraction strengths of RF and the deep learning capabilities of ANN, resulting in a significant improvement in detection accuracy over individual ML models. The results demonstrate the effectiveness of this combined methodology in addressing the evolving landscape of cybersecurity threats.