ZhangLixin0714/fin-risk-models

What it solves

This project provides a library of modular models designed to detect financial risks, fraud, and regulatory compliance failures. It addresses the challenge of identifying complex patterns of financial misconduct—such as money laundering, straw-buyer rings, and "window dressing"—that are often hidden within large transactional datasets.

How it works

The repository consists of a collection of ready-to-use models implemented via SQL logic or algorithms. Each model is focused on a specific risk scenario (e.g., flagging borrowers who share the same recipient account or detecting loans that are repaid immediately after a quarter-end to manipulate financial reports). Users can clone specific model folders to access the core logic and accompanying documentation for implementation.

Who it’s for

This is primarily for risk managers, fraud analysts, and regulatory compliance officers in the financial sector who need practical, pre-defined logic to monitor asset quality and detect anomalies in lending and corporate finance.

Highlights

• Diverse Risk Coverage: Includes models for anti-money laundering (AML), credit risk concentration, and employee conflict-of-interest detection.
• Fraud Detection: Specifically targets organized fraud, such as multiple second-hand mortgage payouts to a single payee.
• Regulatory Compliance: Monitors for non-compliant entrusted payments and overdue investigation responses.
• Modular Design: Each model is self-contained with its own logic, sample use cases, and adjustable parameters.