Banking Fraud Detector

Project Information

Banking Fraud occurs in many ways, some use cases are included below -

• Money Laundering
• Tax Evasion
• Credit Card Fraud.

While these are carried out in extraordinarily different ways by criminals, we believe that patterns exist in most of them. These patterns may not be visible when data is stored in a relational rather linear database , but can be captured when data is stored differently. Thus we want to create a knowledge graph , that focusses on the customers relationships and activities rather than just storing it as another record. This knowledge graph in turn would help match patterns of already existing frauds , thus allowing the bank to take better precautionary measures. We attempt to use various graph analytics methodologies to look for patterns

Technologies Used

Neo4j : Graph Database

Python : Programming Language

CYPHER : Graph Query Language

Flask : Microframework in Python used as a controller in our web application.

MAMP : Server used in order to host our Python application.

Github URL

https://github.com/akashshah59/banking-fraud-detector