Advanced Resampling Techniques and Ensemble Methods for Improved Detection of Imbalanced Healthcare Fraud Cases

Lopes, Snovy Simon (2024) Advanced Resampling Techniques and Ensemble Methods for Improved Detection of Imbalanced Healthcare Fraud Cases. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (750kB) | Preview
Preview	PDF (Configuration Manual) Download (1MB) | Preview

Abstract

Healthcare fraud detection is crucial due to the large amount of financial losses that occur due to it varying from 3% to 10% of the total income or 19 to 65 billion USD a year in the U. S Medicare system. Traditional rule-based fraud detection systems has proven to be insufficient in the current dynamic environment of fraud. However, machine learning (ML) provides potential solutions for designing more effective detection algorithms which is especially noticeable in work with large datasets. But the challenge of imbalanced datasets persists. This study examines the SMOTE-ENN hybrid resampling method with other superior methods and then compares the accuracy of the ensemble methods with the single models. In this study, Random Forest with SMOTE-ENN is compared with the other methods and evaluated using metrics like precision, recall, F1 score, AUC-PR curve, AUC-ROC curve, confusion matrix, etc. The resulting conclusion is SMOTE-ENN with Random Forest performing better compared to the other combinations.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Muntean, Cristina Hava UNSPECIFIED
Uncontrolled Keywords:	SMOTE-ENN; Random forest; Machine learning; Class imbalance; Healthcare fraud; Ensemble methods; Resampling techniques
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science R Medicine > Healthcare Industry Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	03 Sep 2025 11:30
Last Modified:	03 Sep 2025 11:30
URI:	https://norma.ncirl.ie/id/eprint/8737

Actions (login required)

View Item