Machine Learning Approach for Fraud Detection in a Financial Services Application

Rahul Roy Devarakonda

Machine Learning Approach for Fraud Detection in a Financial Services Application

Author(s)	Rahul Roy Devarakonda
Country	United States
Abstract	Fraud detection in financial services is challenging because of the potential sophistication with which fraudulent activities can be carried out. Traditional rule-based detection systems are incapable of adapting to unwarranted changes in the patterns of fraud, which leads to high false positives and undetected fraudulent transactions. This paper proposes a Hybrid ML-Based Fraud Detection Framework wherein unsupervised anomaly detection; supervised classification and adaptive learning might increase the precision of fraud detection while limiting false alarms. Anomaly detection is done through Isolation Forest and Autoencoders, followed by Random Forest, XGBoost, and LSTM, followed by classification of fraud. Experimental results show that XGBoost produced a high accuracy of 97%, while LSTM had the best recall of 95%, distinguishing sequential fraud activities. The hybrid approach strikes a balance between precision and recall thereby ensuring accurate fraud identification with minimal disruption to genuine transactions. This research suggests that the ability to combine different machine learning techniques may help develop a generic, adaptive, and high-performance fraud protection system running in real time for financial applications.
Keywords	Fraud Detection, Machine Learning, Anomaly Detection, Financial Security, Predictive Analytics, Transaction Monitoring
Field	Engineering
Published In	Volume 14, Issue 1, January-March 2023
Published On	2023-01-07
Cite This	Machine Learning Approach for Fraud Detection in a Financial Services Application - Rahul Roy Devarakonda - IJSAT Volume 14, Issue 1, January-March 2023.

View / Download PDF File

About IJSAT Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us	Message on WhatsApp	+91-9687-182-185	editor@ijsat.org

International Journal on Science and Technology

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

Machine Learning Approach for Fraud Detection in a Financial Services Application

Share this