Utilizing Anomaly Detection Methods for Identifying Fraudulent Activities in Credit Card Transactions
N. Siva Rama Krishna Prasad, Amit Gangopadhyay, R. Prabakaran, Nilamadhab Mishra, D. Shyam Prasad, Sneha Kapoor · 2024
Anomaly detection involves identifying unusual occurrences that could signal potential issues, such as security breaches, system failures, financial fraud, structural defects, or medical errors. In the context of digital finance, credit card fraud has become a significant concern, with fraudulent transactions often designed to appear legitimate. To address this, we employ the Synthetic Minority Oversampling Technique (SMOTE) to balance the dataset during sampling, and use the ARIMA model for time series forecasting to ensure accurate results. This research explores various fraud detection techniques and models, including K-Nearest Neighbors (KNN), Random Forest, Support Vector Classifier (SVC), and Decision Trees. We evaluate and compare the performance of these models to determine which offers the highest accuracy in detecting fraud.