Feature Set Generation from Customer Transaction Data for Customer Classification

Verafin is an information technology company that specializes in customer intelligence solutions for small and midsize financial institutions. There is a high demand for automated fraud and money laundering detection and prevention systems since such activities cost millions to the financial industry every year. A key problem with detection techniques is the accurate and descriptive profiling of the account. Thus, it is important to identify the salient features in transaction data that would enable the company to accurately profile the accounts. Through accurate profiling of customer accounts, the search for fraudulent activities can be narrowed down to high-risk groups of accounts. Thus, the intern intends to develop the algorithms to identify a feature set from the account data.

Faculty Supervisor:

Dr. George K.I. Mann

Student:

Dilan Amarasinghe

Partner:

Verafin Inc.

Discipline:

Engineering

Sector:

Finance, insurance and business

University:

Memorial University of Newfoundland