Financial Fraud Detection and Comparison Using Different Machine Learning Techniques
Pratishank Shukla, Mukul Aggarwal, Prakarsh Jain, Parijat Khanna, Madhur Kumar Rana
Abstract
More than ever before, fraudsters are actively targeting financial transactions. This research paper examines the effectiveness of various machine learning techniques in detecting and preventing financial fraud arising due to transactions. In this paper we have compared and analysed 6 different kinds of Machine Learning Techniques i.e. (Naive Bayes, Neural Network, Decision Tree, Support Vector Machine, Logistic Regression, Random Forest) with Random Forest being the most suitable for predicting fraudulent transactions. The research also identifies patterns in fraud cases, as the timing of occurrences and the demographics targeted. The study concludes by suggesting future research directions, including exploring advanced ensemble learning methods, incorporating deep learning algorithms, addressing imbalanced datasets, implementing real-time fraud detection, and extending the research to other sectors such as health. Overall, this study helps in understanding of credit card fraud detection and provides valuable insights for future research.