Network Intrusion Detection System (NIDS) project applies a data-driven methodology using machine learning to detect and classify network intrusions. Network traffic data, stored in CSV format, is used for both training and testing in IBM watsonx.ai Studio. AutoAI automates model selection, tuning, and evaluation, ensuring optimal performance for real-time threat detection.
Approach Highlights Data-Driven Pipeline – From CSV ingestion to model deployment using IBM watsonx.ai and Watson Machine Learning. Automated Model Optimization – AutoAI selects, tunes, and ranks algorithms for best accuracy. Scalable Deployment – REST API integration (future) for live classification and monitoring.
Methodology Data Preparation – Load and preprocess network traffic CSV files. Model Building – Use AutoAI to train, evaluate, and select the best-performing model. Deployment & Monitoring – Deploy via Watson Machine Learning, integrate into NIDS workflow, and monitor performance.