Project Team
Students
Aemen Ali
Computer Science
Penn State Lehigh Valley, Penn State University Park
Faculty Mentors
Richard Martin
Penn State Lehigh Valley
Computer Science, Information Sciences and Technology, Mathematics, and Philosophy
Project
Project Video
Project Abstract
This study aims to develop expert system-based software to determine Fair Lending Risk Assessment (FLRA) risk factors in financial institutions.
Fair Lending regulations mandate banks to provide everyone with equal access to financing possibilities irrespective of the borrower’s race, color, national origin, religion, sex, family position, or disability. There are three recognized types of fair lending discrimination: overt discrimination, disparate impact, and disparate treatment. Several U.S. regulations – such as the Equal Credit Opportunity Act (ECOA), the Community Reinvestment Act (CRA), and the Fair Housing Act (FHA) – protect consumers by prohibiting unfair and discriminatory practices and ensure that commercial banks and saving associations are consistent with safe and sound operations.
The design for the Prolog expert system software was developed by capturing the expertise of fair lending regulations by compiling the Inherent Risk Questionnaire (IRQ) and Controls ratings. The software prompts the user to answer a series of yes and no questions specifically categorized by fundamental risk drivers, such as derogatory remarks, potential customer harm, customer allegations of discrimination, etc. Based on the IRQ and the user’s input, the software computes a risk score and determines the appropriate rating. The Prolog software was tested by feeding sample representative data with varying degrees of response.
The knowledge of fair lending experts captured in this database will help banks identify, measure, control, monitor, and mitigate fair lending risk. Combined with a sound Compliance Management Program, banks will be able to commit to the importance of fair lending regulations, resulting in economic development for underprivileged communities. The qualitative risk rating methodology used by the expert system is subjective and is based on individual responses to the IRQ. The initial version of the knowledge database will not be able to cover all subtle and complex scenarios, and it is expected to evolve in its comprehensiveness towards becoming a mature FLRA expert. Future research will integrate the knowledge base of this expert system with the data analytics of the FLRA, which will help minimize the subjectivity of human response and achieve objective insight into the Fair Lending practices of banks.
Evaluate this Project
Use this form link to provide feedback to the presenters, and add your project evaluation for award(s) consideration.