SaibhargaviMalkapuram/PriviNetra-AI-Based-Privilege-Threat-Detection-Framework

# Privilege Escalation Attack Detection and Mitigation in Cloud Using Machine Learning ## 📌 Overview This project is an AI-based cloud security system developed to detect privilege escalation and insider attacks in cloud environments using Machine Learning algorithms. The system analyzes cloud activity logs and identifies suspicious behavior using ensemble learning techniques. ## 🚀 Features * Insider threat detection * Privilege escalation attack detection * Machine learning-based classification * Performance comparison of algorithms * Confusion matrix visualization * Attack prediction system * GUI-based implementation ## 🛠️ Technologies Used * Python * Machine Learning * Tkinter * Scikit-learn * XGBoost * LightGBM * CatBoost * Pandas * NumPy * Matplotlib * Seaborn ## 📂 Dataset * CERT Insider Threat Dataset Dataset Link: https://github.com/lcd-dal/feature-extraction-for-CERT-insider-threat-test-datasets ## ⚙️ Algorithms Used * Random Forest * AdaBoost * XGBoost * LightGBM * CatBoost ## 📊 Modules 1. Upload CERT Dataset 2. Preprocess & Split Dataset 3. Run Random Forest 4. Run AdaBoost 5. Run XGBoost 6. Run LightGBM 7. Run CatBoost 8. Comparison Graph 9. Attack Prediction ## 💻 System Requirements ### Hardware * Intel i3 Processor or above * 4GB RAM minimum ### Software * Windows OS * Python 3.7+ ## 📦 Installation pip install -r requirements.txt ## ▶️ Run the Project python Main.py OR Run: run.bat ## 📈 Results | Algorithm | Accuracy | | ------------- | -------- | | Random Forest | 86% | | AdaBoost | 88% | | XGBoost | 88.27% | | LightGBM | 97% | ## 🔐 Application Areas * Cloud Security * Cybersecurity * Insider Threat Detection * Enterprise Security Systems ## 📷 Output Add screenshots of: * Dataset Upload * Training Output * Confusion Matrix * Accuracy Graph * Prediction Results ## 📄 License This project is developed for educational and research purposes.