article · IEEE Access
Detecting healthcare fraud presents major difficulties due to severe class imbalance, which can cause standard machine learning models to overfit, lose critical data, or incorporate noise. To address this in the Medicare Part B dataset, a novel approach extracts the provider type feature and synthesises new instances through random replication, increasing minority class diversity. Following this, the hybrid resampling method SMOTE-ENN combines synthetic oversampling with edited nearest neighbours to remove noisy samples and balance the dataset. Six machine learning models were tested using metrics such as precision, recall, F1 score, AUC-ROC, and area under the precision-recall curve. The experiments demonstrated that the Decision Trees classifier achieved superior results, attaining a score of 0.99 across all evaluated metrics.
Healthcare fraud imposes heavy financial and operational costs, but fraudulent claims are rare compared to legitimate ones. Traditional machine learning models struggle with this disparity, leading to missed fraud or high false alarms. Developing resampling methods that remove noise while amplifying rare fraud cases helps automated systems identify illicit claims much more accurately, protecting vital medical resources and public expenditure.
This approach is targeted at healthcare fraud detection systems, specifically relevant to administrators, health insurers, and Medicare compliance auditors managing imbalanced claims data. Tested on Medicare Part B records with six machine learning classifiers, the method operates at an applied research stage. Further integration would require adapting the pipeline to live claims processing software and validating performance against streaming, multi-provider administrative data streams.
AI-generated from the published abstract. Always read the original work before citing.
The healthcare fraud detection field is constantly evolving and faces significant challenges, particularly when addressing imbalanced data issues. Previous studies mainly focused on traditional machine learning (ML) techniques, often struggling with imbalanced data. This problem arises in various aspects. It includes the risk of overfitting with Random Oversampling (ROS), noise introduction by the Synthetic Minority Oversampling Technique (SMOTE), and potential crucial information loss with Random Undersampling (RUS). Moreover, improving model performance, exploring hybrid resampling techniques, and enhancing evaluation metrics are crucial for achieving higher accuracy with imbalanced datasets. In this paper, we present a novel approach to tackle the issue of imbalanced datasets in healthcare fraud detection, with a specific focus on the Medicare Part B dataset. First, we carefully extract the categorical feature "Provider Type" from the dataset. This allows us to generate new, synthetic instances by randomly replicating existing types, thereby increasing the diversity within the minority class. Then, we apply a hybrid resampling method named SMOTE-ENN, which combines the Synthetic Minority Over-sampling Technique (SMOTE) with Edited Nearest Neighbors (ENN). This method aims to balance the dataset by generating synthetic samples and removing noisy data to improve the accuracy of the models. We use six machine learning (ML) models to categorize the instances. When evaluating performance, we rely on common metrics like accuracy, F1 score, recall, precision, and the AUC-ROC curve. We highlight the significance of the Area Under the Precision-Recall Curve (AUPRC) for assessing performance in imbalanced dataset scenarios. The experiments show that Decision Trees (DT) outperformed all the classifiers, achieving a score of 0.99 across all metrics.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1109/access.2024.3385781
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
