The increasing dependence on digital communication and internet-based services has made computer networks more vulnerable to cyber threats. Malicious activities such as denial-of-service attacks, port scanning, brute-force attempts, and unauthorized access continue to challenge the security of modern network infrastructures. Conventional intrusion detection techniques often rely on predefined signatures, making them less effective against newly emerging attack patterns. To address this limitation, this study proposes an intelligent network intrusion detection system based on ensemble machine learning techniques. The proposed approach combines multiple classification algorithms to improve the accuracy and reliability of intrusion detection while reducing false alarms. Network traffic data are analysed using selected features that represent normal and malicious behaviour, enabling the system to identify different categories of attacks efficiently. By integrating the strengths of multiple machine learning models, the proposed system achieves more consistent detection performance than individual classifiers. The study demonstrates that ensemble learning can significantly enhance network security and provides an effective framework for developing intelligent intrusion detection solutions suitable for real-world cybersecurity environments.

Keywords— Network Intrusion Detection System, Cybersecurity, Ensemble Machine Learning, Network Traffic Analysis, Random Forest, XGBoost, LightGBM, Attack Detection, Artificial Intelligence, Network Security. .