How hirdeshpal888527/graphsentinel works
GraphSentinel combines transaction-level fraud scoring, account-network analysis, interpretable rules and GraphSAGE experiments, then serves results through a FastAPI API and React investigation dashboard. The graph preserves the documented analytical stages and dashboard capabilities; several training and data-preparation links are documented in the README rather than shown in sampled implementation excerpts.
Main entry points: App.jsx Dashboard shell · routes.py API routes · services.py Prediction services · data_processor.py Feature engineering
Components 24
Fraud analysis
- Data preparation
src/data_cleaner.py - Feature engineering
src/data_processor.py - Transaction ML
src/train_ml_model.py - Account graph
src/graph_builder.py - Graph analysis
src/advanced_graph_analysis.py - Rule detection
src/fraud_detector.py - GraphSAGE learning
src/train_gnn_model.py
API and artifacts
- Model artifacts
models - Analysis outputs
data/processed - FastAPI app
backend/app.py - API routes
backend/routes.py - Request schemas
backend/schemas.py - Prediction services
backend/services.py
Analyst dashboard
- Dashboard shell
frontend/src/App.jsx - Transaction scoring
frontend/src/pages/TransactionScoring.jsx - Risk queue
frontend/src/pages/RiskQueue.jsx - Transaction detail
frontend/src/components/TransactionDetail.jsx - Case review
frontend/src/pages/CaseReview.jsx - Graph intelligence
frontend/src/pages/GraphIntelligence.jsx - Model performance
frontend/src/pages/ModelPerformance.jsx - Rules and signals
frontend/src/pages/RulesSignals.jsx - Overview
frontend/src/pages/Overview.jsx
Other components
- Analyst
- Transaction data
Connections 27
- Transaction data → Data preparation: prepares
README.md - Data preparation → Feature engineering: engineers features
README.md - Feature engineering → Transaction ML: trains
README.md - Transaction data → Account graph: builds network
README.md - Account graph → Graph analysis: analyzes structure
README.md - Feature engineering → GraphSAGE learning: supports learning
README.md - Transaction ML → Model artifacts: produces model
- GraphSAGE learning → Model artifacts: produces model
- Graph analysis → Rule detection: informs risk
README.md - Transaction data → Rule detection: scores signals
README.md - Model artifacts → Prediction services: loads models
backend/services.py - Analysis outputs → Prediction services: loads metrics
backend/services.py - FastAPI app → API routes: registers routes
backend/app.py - FastAPI app → Prediction services: preloads artifacts
backend/app.py - API routes → Request schemas: validates requests
backend/routes.py - API routes → Prediction services: calls prediction
backend/routes.py - Transaction scoring → API routes: requests scoring
README.md - Dashboard shell → Transaction scoring: renders
frontend/src/App.jsx - Dashboard shell → Risk queue: renders
frontend/src/App.jsx - Dashboard shell → Graph intelligence: renders
frontend/src/App.jsx - Dashboard shell → Model performance: renders
frontend/src/App.jsx - Dashboard shell → Rules and signals: renders
frontend/src/App.jsx - Dashboard shell → Case review: renders
frontend/src/App.jsx - Dashboard shell → Overview: renders
frontend/src/App.jsx - Risk queue → Transaction detail: opens details
frontend/src/pages/RiskQueue.jsx - Analyst → Dashboard shell: investigates
README.md - Analyst → Risk queue: reviews risk
README.md