NetSentinelAI – AI-Powered Network Intrusion Detection System
Intelligent network threat detection using machine learning.
NetSentinelAI is an AI-powered network intrusion detection system designed to identify malicious network traffic and visualize security threats. It uses a Random Forest machine-learning model trained on the CICIDS2017 dataset to classify benign traffic and multiple cyberattack categories. The platform includes an interactive Flask dashboard with attack alerts, analytics, attacker information, CSV-based detection, and IP-blocking functionality.
Technology stack
Problem
Traditional signature-based intrusion detection systems can struggle to identify evolving attack patterns. Security teams also need an efficient way to analyze network traffic, identify malicious behavior, and monitor threats through a centralized dashboard.
Approach
Developed a machine-learning classification pipeline using CICIDS2017 network traffic data. Processed approximately 2.83 million network flows with 78 features and trained a Random Forest classifier to identify 15 traffic classes. Integrated the prediction engine with a Flask backend and interactive frontend dashboard.
Solution
Built an end-to-end network intrusion detection prototype that classifies network traffic and displays security alerts, attack timelines, and top attackers. Added CSV upload and detection functionality along with simulated or configurable firewall-based IP-blocking controls.
Results
The project README reports approximately 96% overall accuracy and a macro F1-score of 0.82. Evaluation covers approximately 2.83 million network flows. The results also reveal weaker performance on rare attack categories, providing a clear direction for improving minority-class detection.