Abstract :

Medicare Fraud Poses A Significant Challenge To Healthcare Systems By Increasing Operational Costs, Misallocating Medical Resources, And Compromising The Integrity Of Healthcare Services. Fraudulent Activities Such As False Claims, Duplicate Billing, Unnecessary Medical Procedures, And Identity Misuse Result In Substantial Financial Losses For Government And Healthcare Organizations. Detecting Medicare Fraud Is Particularly Difficult Because Healthcare Datasets Are Highly Imbalanced, With Fraudulent Cases Representing Only A Small Fraction Of Total Claims. Traditional Machine Learning Models Often Struggle To Accurately Identify These Minorityclass Instances, Leading To Poor Fraud Detection Performance. This Paper Presents A Machine Learning-based Medicare Fraud Detection Framework That Addresses Class Imbalance Using The Synthetic Minority Oversampling Technique– Edited Nearest Neighbor (SMOTE-ENN) Method. The Proposed Approach Combines Data Balancing And Noise Reduction Techniques To Improve The Representation Of Fraudulent Claims Within The Dataset. Machine Learning Algorithms Are Then Applied To Classify Healthcare Claims As Legitimate Or Fraudulent Based On Billing Patterns, Patient Records, And Provider Behaviors. Experimental Analysis Demonstrates That The Integration Of SMOTE-ENN Significantly Enhances Classification Accuracy, Recall, And Fraud Detection Performance While Reducing False Negative Rates. The Proposed Framework Provides An Effective, Scalable, And Intelligent Solution For Improving Medicare Fraud Detection And Healthcare Financial Security.

Published:

26-9-2026

Issue:

Vol. 26 No. 9 (2026)

Page Nos:

448-458

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite