Graph-based anomaly detection (GBAD) approaches are among the most popular techniques used to analyze connectivity patterns in communication networks and identify suspicious behaviors. Given the different GBAD approaches proposed for fraud detection, in this study, we develop a framework to synthesize the existing literature on the application of GBAD methods in fraud detection published between 2007 and 2018. This study aims to investigate the present trends and identify the key challenges that require significant research efforts to increase the credibility of the technique. Additionally, we provide some recommendations to deal with these challenges.

Fraud detection: A systematic literature review of graph-based anomaly detection approaches

In the United States, advances in technology and medical sciences continue to improve the general well-being of the population. With this continued progress, programs such as Medicare are needed to help manage the high costs associated with quality healthcare. Unfortunately, there are individuals who commit fraud for nefarious reasons and personal gain, limiting Medicare’s ability to effectively provide for the healthcare needs of the elderly and other qualifying people. To minimize fraudulent activities, the Centers for Medicare and Medicaid Services (CMS) released a number of “Big Data” datasets for different parts of the Medicare program. In this paper, we focus on the detection of Medicare fraud using the following CMS datasets: (1) Medicare Provider Utilization and Payment Data: Physician and Other Supplier (Part B), (2) Medicare Provider Utilization and Payment Data: Part D Prescriber (Part D), and (3) Medicare Provider Utilization and Payment Data: Referring Durable Medical Equipment, Prosthetics, Orthotics and Supplies (DMEPOS). Additionally, we create a fourth dataset which is a combination of the three primary datasets. We discuss data processing for all four datasets and the mapping of real-world provider fraud labels using the List of Excluded Individuals and Entities (LEIE) from the Office of the Inspector General. Our exploratory analysis on Medicare fraud detection involves building and assessing three learners on each dataset. Based on the Area under the Receiver Operating Characteristic (ROC) Curve performance metric, our results show that the Combined dataset with the Logistic Regression (LR) learner yielded the best overall score at 0.816, closely followed by the Part B dataset with LR at 0.805. Overall, the Combined and Part B datasets produced the best fraud detection performance with no statistical difference between these datasets, over all the learners. Therefore, based on our results and the assumption that there is no way to know within which part of Medicare a physician will commit fraud, we suggest using the Combined dataset for detecting fraudulent behavior when a physician has submitted payments through any or all Medicare parts evaluated in our study.

/pdf/big-data-fraud-detection-using-multiple-medicare-data-23445fyobc.pdf

Big Data fraud detection using multiple medicare data sources

Access to affordable healthcare is a nationwide concern that impacts a large majority of the United States population Medicare is a Federal Government healthcare program that provides affordable health insurance to the elderly population and individuals with select disabilities Unfortunately, there is a significant amount of fraud, waste, and abuse within the Medicare system that costs taxpayers billions of dollars and puts beneficiaries’ health and welfare at risk Previous work has shown that publicly available Medicare claims data can be leveraged to construct machine learning models capable of automating fraud detection, but challenges associated with class-imbalanced big data hinder performance With a minority class size of 003% and an opportunity to improve existing results, we use the Medicare fraud detection task to compare six deep learning methods designed to address the class imbalance problem Data-level techniques used in this study include random over-sampling (ROS), random under-sampling (RUS), and a hybrid ROS–RUS The algorithm-level techniques evaluated include a cost-sensitive loss function, the Focal Loss, and the Mean False Error Loss A range of class ratios are tested by varying sample rates and desirable class-wise performance is achieved by identifying optimal decision thresholds for each model Neural networks are evaluated on a 20% holdout test set, and results are reported using the area under the receiver operating characteristic curve (AUC) Results show that ROS and ROS–RUS perform significantly better than baseline and algorithm-level methods with average AUC scores of 08505 and 08509, while ROS–RUS maximizes efficiency with a 4× speedup in training time Plain RUS outperforms baseline methods with up to 30× improvements in training time, and all algorithm-level methods are found to produce more stable decision boundaries than baseline methods Thresholding results suggest that the decision threshold always be optimized using a validation set, as we observe a strong linear relationship between the minority class size and the optimal threshold To the best of our knowledge, this is the first study to compare multiple data-level and algorithm-level deep learning methods across a range of class distributions Additional contributions include a unique analysis of the relationship between minority class size and optimal decision threshold and state-of-the-art performance on the given Medicare fraud detection task

/pdf/medicare-fraud-detection-using-neural-networks-12q9mdfu2l.pdf

Medicare fraud detection using neural networks

Healthcare is an integral component in people’s lives, especially for the rising elderly population, and must be affordable. Medicare is one such healthcare program. Claims fraud is a major contributor to increased healthcare costs, but its impact can be lessened through fraud detection. In this paper, we compare several machine learning methods to detect Medicare fraud. We perform a comparative study with supervised, unsupervised, and hybrid machine learning approaches using four performance metrics and class imbalance reduction via oversampling and an 80-20 undersampling method. We group the 2015 Medicare data into provider types, with fraud labels from the List of Excluded Individuals/Entities database. Our results show that the successful detection of fraudulent providers is possible, with the 80-20 sampling method demonstrating the best performance across the learners. Furthermore, supervised methods performed better than unsupervised or hybrid methods, but these results varied based on the class imbalance sampling technique and provider type.

Medicare Fraud Detection Using Machine Learning Methods

Healthcare in the United States is a critical aspect of most people's lives, particularly for the aging demographic. This rising elderly population continues to demand more cost-effective healthcare programs. Medicare is a vital program serving the needs of the elderly in the United States. The growing number of Medicare beneficiaries, along with the enormous volume of money in the healthcare industry, increases the appeal for, and risk of, fraud. In this paper, we focus on the detection of Medicare Part B provider fraud which involves fraudulent activities, such as patient abuse or neglect and billing for services not rendered, perpetrated by providers and other entities who have been excluded from participating in Federal healthcare programs. We discuss Part B data processing and describe a unique process for mapping fraud labels with known fraudulent providers. The labeled big dataset is highly imbalanced with a very limited number of fraud instances. In order to combat this class imbalance, we generate seven class distributions and assess the behavior and fraud detection performance of six different machine learning methods. Our results show that RF100 using a 90:10 class distribution is the best learner with a 0.87302 AUC. Moreover, learner behavior with the 50:50 balanced class distribution is similar to more imbalanced distributions which keep more of the original data. Based on the performance and significance testing results, we posit that retaining more of the majority class information leads to better Medicare Part B fraud detection performance over the balanced datasets across the majority of learners.

/pdf/the-effects-of-varying-class-distribution-on-learner-3mgplhg24n.pdf

The effects of varying class distribution on learner behavior for medicare fraud detection with imbalanced big data

This paper presents a novel approach to estimating healthcare fraud (HCF) risk that applies network algorithms to graphs derived from open source datasets. One group of algorithms calculates behavioral similarity to known fraudulent and non-fraudulent healthcare providers with respect to measurable healthcare activities, such as medical procedures and drug prescriptions. Another set of algorithms estimates propagation of risk from fraudulent healthcare providers through geospatial collocation, i.e., shared practice locations or other addresses. The algorithms were evaluated with respect to their ability to predict a provider's presence on the Office of the Inspector General's list of providers excluded from participation in Medicare and other Federal healthcare programs (exclusion). In an empirical evaluation, a combination of 11 features achieved an f-score of 0.919 and a ROC area of 0.960 in exclusion prediction. An ablation analysis showed that most of this predictive accuracy was the result of features that measure risk propagation through geospatial collocation.

Jeffrey Gold

Papers

Graph analytics for healthcare fraud risk estimation