Classification for Imbalanced Credit Card Transaction Data

Faculty Mentor

Dan Li

Presentation Type

Poster

Start Date

5-8-2024 11:15 AM

End Date

5-8-2024 1:00 PM

Location

PUB NCR

Primary Discipline of Presentation

Computer Science

Abstract

It is important for the credit card companies to identify fraudulent transactions to avoid financial loss for the customers and the companies. The challenge of fraud detection lies in the imbalanced feature of transaction data which makes traditional classification algorithms infeasible. This research investigates the methodologies that are commonly employed to deal with imbalanced datasets. Specifically, over-sampling, under-sampling, and Synthetic Minority Over-sample Technique (SMOTE) are studied. In addition, to better understand the statistical features of the credit card transaction data, we implement a stream mining algorithm, DGIM, to analyze the occurrences of fraudulent transactions over sliding windows. This study contributes to the understanding of effective strategies for mitigating imbalanced datasets and enhancing fraud detection mechanisms in financial system.

This document is currently not available here.

Share

COinS
 
May 8th, 11:15 AM May 8th, 1:00 PM

Classification for Imbalanced Credit Card Transaction Data

PUB NCR

It is important for the credit card companies to identify fraudulent transactions to avoid financial loss for the customers and the companies. The challenge of fraud detection lies in the imbalanced feature of transaction data which makes traditional classification algorithms infeasible. This research investigates the methodologies that are commonly employed to deal with imbalanced datasets. Specifically, over-sampling, under-sampling, and Synthetic Minority Over-sample Technique (SMOTE) are studied. In addition, to better understand the statistical features of the credit card transaction data, we implement a stream mining algorithm, DGIM, to analyze the occurrences of fraudulent transactions over sliding windows. This study contributes to the understanding of effective strategies for mitigating imbalanced datasets and enhancing fraud detection mechanisms in financial system.