User Tools

Site Tools


7a.020.UVA - Deep Learning and Adversarial Learning in Credit Card Fraud

Project - Summary

Credit card fraud is a problem that can cost banks billions of dollars annually, leading to increased incentives among financial institutions for the development of fast, effective and dynamic fraud detection systems. This research project addresses credit card fraud detection through a semi-supervised approach, in which clusters of account profiles are created and used for modeling classifiers. Accounts are profiled based on their behavioral trends and clustered into similar groups. Groups are further identified as distinct customer segments based on purchase characteristics such as amount, frequency or distance. Random forest and XGBoost classifiers are trained on an entire sample and compared against classifiers trained at the transaction level across each cluster. This research concludes that the overall weighted performance of classifiers trained at the cluster level does not significantly outperform classifiers trained on the full sample. However, this research finds that clustering can be used to find meaningful groups of account holders that also have varying fraud rates across each cluster. Additionally, some classifiers trained on specific clusters yield significant improvements in performance over the baseline, whereas classifiers for other clusters do not perform as well as the baseline. This research also concludes that the optimal classifier for a given cluster varies by cluster, highlighting the potential for further development of new classifiers which may perform well on clusters that currently exhibit underperforming models.

Project - Team

Team Member Role Email Phone Number Academic Site/IAB
Stephen Adams PI (757) 870-4954 University of Virginia
Navin Kasa Capstone Team Member Not Available Not Available University of Virginia
Andrew Dahbura Capstone Team Member Not Available Not Available University of Virginia
Charishma Ravoori Capstone Team Member Not Available Not Available University of Virginia
David Lutz Project Mentor Not Available Not Available Capital One

Project - Novelty of Approach

This project will be a continuation of the prior work and further develop the concepts.

Project - Deliverables

1 Code for use at Capital One
2 Research summaries and presentations
3 SIEDS papers

Project - Benefits to IAB

This research is a first step in demonstrating the usefulness of unsupervised methods in credit card fraud. The results are beneficial to any member of the banking industry. The process of clustering large data sets and building a diverse set of classifiers custom to each cluster could be expanded to other industries such as healthcare.

Project - Presentation Video

Project - Documents

FilenameFilesizeLast modified
7a.020.uva_final_report.docx941.2 KiB2019/08/19 12:43
7a.020.uva_quad_chart_2018_spring_meeting.pptx590.4 KiB2019/08/13 15:03
7a.020.uva_year_7_cvdi_mid-year_report.docx237.3 KiB2019/08/13 15:03
7a.020.uva_year_7_cvdi_mid-year_report_next_version_or_duplicate.docx237.4 KiB2019/08/13 15:03
7a.020.uva_2018_fall_meeting_poster.pptx469.4 KiB2019/08/13 15:03
7a.020.uva_executive_summary.docx51.0 KiB2019/08/13 15:03
7a.020.uva_confluence_project_page.pdf142.3 KiB2019/08/13 15:03
projects/year7/7a.020.uva.txt · Last modified: 2019/08/20 10:39 by sally.johnson