Abstract
<jats:p>This research designs and evaluates a Network Intrusion Detection System (NIDS) using machine learning, focusing on Decision Tree Classifiers to meet growing cybersecurity demands. The system follows a two-stage process: an anomaly detection model first flags unusual network traffic, then an attack classification model categorizes it into types like DoS, Probe, R2L, and U2R. The approach involves data preprocessing, training, and evaluation using metrics such as accuracy, precision, recall, and F1-score. Results show machine learning's promise in automating intrusion detection and classification. The study notes current limitations and suggests future work, including real-time deployment, advanced models, and testing across diverse network environments to build more adaptive, scalable, and effective security solutions.</jats:p>