ABSTRACT With the rapid growth of internet connectivity and digital infrastructure, network security has become a major concern for organizations and individuals. Cyber-attacks such as denial-of-service attacks, malware infections, and unauthorized access attempts pose significant threats to computer networks. Traditional intrusion detection systems rely on rule-based mechanisms that detect known attack patterns but often fail to identify new or evolving cyber threats. Machine learning techniques provide an effective approach for detecting network intrusions by analyzing large volumes of network traffic data and identifying abnormal behavior patterns. This study proposes a machine learning-based intrusion detection framework using classification algorithms such as Decision Tree, Support Vector Machine, Random Forest, and Gradient Boosting. Network traffic features including protocol type, connection duration, and packet size are analyzed to classify normal and malicious network activities. The performance of the proposed models is evaluated using Accuracy, Precision, Recall, and F1-Score metrics. Experimental results demonstrate that ensemble learning algorithms achieve higher detection accuracy and can effectively improve cybersecurity systems. Key words: Network Security, Intrusion Detection System, Machine Learning, Cybersecurity Analytics, Network Traffic Classification