Comparative study between ML approaches in Intrusion Detection Context

In today's rapidly advancing technological landscape, just like any other field, attackers are also constantly evolving their tactics, highlighting the importance of developing novel security tools to keep pace with these dynamic threats. Machine learning and deep learning algorithms have emerged as powerful tools in the field of cybersecurity. They offer the advantage of being able to adapt and learn from patterns and anomalies in network data, enabling them to effectively detect, prevent, and mitigate cyber attacks without relying solely on predefined rules or human intervention. In this context, a comparative study was conducted to evaluate the effectiveness of various machine learning approaches in the context of intrusion detection. The study focused on different algorithms, including Gaussian Naïve-Bayes, Logistic Regression, Decision Tree, Random Forest, K-Nearest Neighbour, AdaBoost and Deep learning algorithms such as Artificial Neural Network (ANN), Convolution Neural Network (CNN), and Recurrent Neural Network (RNN). The study concluded that amongst the various machine learning algorithms are giving an accuracy of over 99% for malware detection inside various test CICIDS2017 dataset.