ML-Based Cyber Intrusion Detection & Deception Platform

Built the ML detection pipeline for a cybersecurity platform combining intrusion detection with honeypot deception. Trained XGBoost classifiers on UNSW-NB15, CICIDS2017, and NSL-KDD for binary, category, and sub-attack classification. Built preprocessing pipelines (encoding, scaling, stratified splitting), handled class imbalance via balanced sample weighting, and evaluated with accuracy, classification reports, and confusion matrices. Detected traffic feeds a deception layer that redirects confirmed attackers to honeypots.