A python-based Network Intusion Detection System, for every one.

Machine Learning - Network Intrusion Detection System

This project implements a modern Network Intrusion Detection System (NIDS) using deep learning and machine learning to identify cybersecurity threats. It works with an intelligent pipeline that includes PCA‑based feature reduction, autoencoder anomaly detection, and classifiers like XGBoost. Built on a cleaned version of the CICIDS2017 dataset

Anomaly-Based Network Intrusion Detection Using Ensemble Learning

Built a dual-model intrusion detection system using the NSL-KDD dataset. The binary model detects normal vs. malicious traffic, while the multiclass model classifies attacks as DoS, Probe, R2L, or U2R. Applied advanced preprocessing, dimensionality reduction, and ensemble methods (XGBoost, LightGBM, SVM) with recall focused cross-validation

Network Intrusion Detection System Using IBM Cloud Lite and Watsonx.ai Studio