Financial Compliance Risk Management Using Knowledge-Based Systems and Analytics

Loading...
Thumbnail Image

Authors

Doost, Alireza Farhang

Date

2025-04-03

Type

thesis

Language

eng

Keyword

Risk Management , Financial Regulatory Compliance , Partially Observable Markov Decision Process , Machine Learning , Knowledge-based Systems , Feature Engineering

Research Projects

Organizational Units

Journal Issue

Alternative Title

Abstract

In financial institutions, risk assessment of client accounts is a sensitive task requiring a blend of manual review and automated rating. While scarcity of labelled data, compliance requirements, and complexity of cases are significant obstacles in complete automation, manual review leverages expert knowledge to address these challenges. This thesis presents two methodologies that distinctly study and improve manual client review process and automated client risk rating. Both methodologies are developed to address real-world cases using real-world data with practical applications in financial institutions. In the first study, we develop a novel knowledge-based decision support framework using a Partially Observable Markov Decision Process (POMDP) to improve the manual client review process in terms of reduction in backlogs, optimal assignment of cases to reviewers, and model interpretability with respect to compliance expectations. The POMDP model enforces a forward-looking sequential decision approach that incorporates operational costs associated with manual review, and potential consequences of misdiagnosis (false positive and false negative) therefore taking into account the significant costs of non-compliance and regulatory penalties. The second study introduces a feature engineering method to improve the accuracy, robustness, and interpretability of automated risk rating classification decisions of a Logistic Regression model used for Client Risk Rating (CRR). The model has been used in production at a major bank.

Description

Citation

Publisher

License

Queen's University's Thesis/Dissertation Non-Exclusive License for Deposit to QSpace and Library and Archives Canada
ProQuest PhD and Master's Theses International Dissemination Agreement
Intellectual Property Guidelines at Queen's University
Copying and Preserving Your Thesis
This publication is made available by the authority of the copyright owner solely for the purpose of private study and research and may not be copied or reproduced except as permitted by the copyright laws without written authority from the copyright owner.
Attribution-NonCommercial 4.0 International

Journal

Volume

Issue

PubMed ID

External DOI

ISSN

EISSN