Fraud Shield: Protecting Transactions with Machine Learning Based Fraud Detection
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
The growing scale of online financial activities has led to a marked surge in credit card fraud incidents, creating substantial obstacles for both financial organizations and individual users. This project presents an effective methodology for credit card fraud detection, grounded in machine learning principles and specifically utilizing Logistic Regression. A key aspect addressed is the significant class imbalance within the dataset-typified by the scarcity of fraudulent transactions relative to legitimate ones-which is proficiently handled through careful data preprocessing and appropriate resampling techniques. A Logistic Regression model is subsequently trained to accurately differentiate between fraudulent and genuine transactions based on carefully selected features. Experimental validation confirms that Logistic Regression achieves high accuracy while sustaining robust precision and recall rates, which are crucial for minimizing false negatives in fraud detection. This model demonstrates that reliable detection performance can be attained with reduced computational complexity using a straightforward yet powerful classifier, thereby making it well-suited for real-time fraud detection systems.
Fields of Study & Subjects
Imbalanced Data Classification Techniques
Privacy-Preserving Technologies in Data
Financial Distress and Bankruptcy Prediction
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 26 March 2026
- Language
- EN