Credit Card Fraud Detection System Demo

System Controls

0
Total Transactions
0
Fraud Transactions
0%
Fraud Rate
$0
Average Amount

Transaction Feed

Transaction Volume

Fraud Detection Rate

System Architecture

graph TB subgraph Data Sources DG[Data Generator] RT[Real Transactions] end subgraph Message Broker K[Kafka] end subgraph Processing PS[PySpark Processor] N[Neo4j Graph DB] end subgraph ML Pipeline GNN[GNN Model] GR[GraphRAG] end subgraph Visualization D[Dashboard] end DG -->|Synthetic Data| K RT -->|Real Data| K K -->|Stream| PS PS -->|Processed Data| N N -->|Graph Data| GNN GNN -->|Predictions| GR GR -->|Enhanced Results| D N -->|Graph Queries| D classDef active fill:#e1f5fe,stroke:#01579b,stroke-width:2px; classDef inactive fill:#f5f5f5,stroke:#9e9e9e,stroke-width:1px; classDef highlight fill:#fff3e0,stroke:#ef6c00,stroke-width:2px;

Data Flow

sequenceDiagram participant DG as Data Generator participant K as Kafka participant PS as PySpark participant N as Neo4j participant GNN as GNN Model participant GR as GraphRAG participant D as Dashboard Note over DG,D: Transaction Processing DG->>K: Publish Transaction K->>PS: Stream Transaction PS->>PS: Feature Engineering PS->>N: Store in Graph DB N->>GNN: Analyze Graph GNN->>GR: Get Predictions GR->>GR: Retrieve Similar Patterns GR->>D: Send Results D->>D: Update Visualization Note over DG,D: Fraud Detection DG->>K: Generate Fraud Pattern K->>PS: Stream Suspicious Data PS->>N: Update Graph N->>GNN: Detect Anomalies GNN->>GR: Analyze Patterns GR->>D: Alert Dashboard D->>D: Highlight Fraud