RIFT 2026 · Graph Theory Track

Money Mule Detection
Graph-Based Financial Crime Engine

Detect circular fund routing, smurfing, and layered shell networks using transaction graphs and behavioral risk scoring.

Upload CSV How It Works
0Transactions
0Suspicious Accounts
0Fraud Rings

Upload Transaction Dataset

Detection Methodology

1

Convert CSV into transaction graph

2

Detect circular fund routing (cycles)

3

Analyze fan-in / fan-out smurfing

4

Assign suspicion scores (0–100)

5

Generate investigator-ready outputs

Suspicious Accounts

Account ID Suspicion Score Detected Pattern

About This Project

Financial crime has evolved beyond simple, rule-based fraud patterns. One of the most sophisticated and damaging techniques used today is money muling, where illicit funds are deliberately moved through networks of accounts to conceal their true origin. These networks often involve circular transfers, layered intermediaries, and high-velocity transaction flows that are extremely difficult to detect using traditional database queries or threshold-based monitoring systems. This project, Money Mule Detection – Graph-Based Financial Crime Engine, was developed as part of the RIFT 2026 Hackathon (Graph Theory / Financial Crime Detection Track) to address this challenge using graph analytics, temporal analysis, and explainable risk scoring.