Real-time intrusion detection with machine learning powered by Android mobile technology.
Continuous packet capture and analysis with <200ms latency for immediate threat detection.
8-class threat detection using Random Forest and Gradient Boosting with >95% accuracy.
Detects DoS/DDoS, PortScan, BruteForce, Botnet, WebAttack, Infiltration and more.
Instant notifications with severity levels and actionable recommendations.
Interactive dashboards showing traffic patterns and threat visualizations.
Combines signature-based detection with adaptive ML models for comprehensive coverage.
Java/Kotlin implementation with Material Design 3 and cyberpunk UI theme
Packet capture library for real-time network traffic analysis
MPAndroidChart for interactive threat analytics dashboards
Java backend with REST APIs for mobile communication
SMILE/Weka for Random Forest, Gradient Boosting, and DL4J models
Trained on CIC-IDS2017 & UNSW-NB15 cybersecurity datasets
Across all threat categories with comprehensive testing
Minimizing unnecessary alerts for operational efficiency
From packet capture to threat classification