How nathesh6/cyber-hacking-breach-detection-predication-using-xg-boost works
This Flask application supports two user workflows: submit URLs for current-threat detection or future-risk prediction, then view results and dashboard history. The backend extracts URL features, applies XGBoost and detection heuristics, and persists scan data in SQLite. Model training is a separate capability that produces the detection and prediction artifacts. The README also describes dataset uploads and classification reasons; the sampled browser code confirms single and batch URL submission, while backend routing and template wiring are only partly shown, so uncertain links are omitted.
Main entry points: app.py Flask API and pages · train_models.py Model training · detection.js Detection UI · prediction.js Prediction UI
Components 16
Web interface
- Demo login
backend/static/js/auth.js - Detection UI
backend/static/js/detection.js - Prediction UI
backend/static/js/prediction.js - Dashboards
backend/static/js/dashboard.js - Browser history
backend/static/js/prediction.js
URL analysis
- Flask API and pages
backend/app.py - URL features
backend/app.py - Threat detection
backend/app.py - Risk prediction
backend/app.py - Threat scoring
backend/app.py
History and stats
- Scan history
backend/security_analytics.db
Model training
- Trained model artifacts
backend/models - Model training
backend/train_models.py - Keyword dataset
backend/suspicious_keywords_500.csv - URL training dataset
backend/url_dataset_100k.csv
Other components
- Security user
Connections 18
- Security user → Demo login: signs in
backend/static/js/auth.js - Security user → Detection UI: submits URLs
backend/static/js/detection.js - Security user → Prediction UI: submits URLs
backend/static/js/prediction.js - Detection UI → Flask API and pages: POST /api/detect
backend/static/js/detection.js - Prediction UI → Flask API and pages: POST /api/predict
backend/static/js/prediction.js - Flask API and pages → URL features: extracts features
backend/app.py - Flask API and pages → Threat detection: dispatches detection
backend/app.py - Flask API and pages → Risk prediction: dispatches prediction
backend/app.py - Threat detection → Threat scoring: scores threats
backend/app.py - Threat detection → Trained model artifacts: uses model
- Risk prediction → Trained model artifacts: uses model
- Flask API and pages → Scan history: reads and writes
backend/app.py - Dashboards → Flask API and pages: loads dashboard data
- Detection UI → Browser history: stores results
backend/static/js/detection.js - Prediction UI → Browser history: stores results
backend/static/js/prediction.js - Model training → URL training dataset: loads dataset
backend/train_models.py - Model training → Keyword dataset: loads keywords
- Model training → Trained model artifacts: saves models
backend/train_models.py