Medicare Fraud

Medicare fraud, the intentional misrepresentation of medical billing to obtain higher reimbursements, represents a significant financial drain on the US healthcare system and similar programs globally. Current research focuses on applying machine learning techniques, particularly deep learning architectures like autoencoders and LSTM networks, often coupled with data preprocessing methods such as SMOTE to address class imbalance issues inherent in fraud detection datasets. These models aim to identify patterns indicative of fraudulent activity within large-scale claims data, offering improved accuracy and explainability compared to traditional methods, thereby informing both fraud prevention strategies and resource allocation.

Papers

November 5, 2022

Unsupervised Machine Learning for Explainable Health Care Fraud Detection
Shubhranshu Shekhar, Jetson Leder-Luis, Leman Akoglu
Service Provider Unsupervised Machine Learning Fraud Pattern Graph Based Fraud Detection Medicare Fraud

June 3, 2022

Impact of the composition of feature extraction and class sampling in medicare fraud detection
Akrity Kumari, Narinder Singh Punn, Sanjay Kumar Sonbhadra, Sonali Agarwal
Machine Learning Global Impact Dimensionality Reduction Feature Extraction Fraud Detection Compositional Ability Medicare Fraud

March 11, 2022

Multiple Inputs Neural Networks for Medicare fraud Detection
Mansour Zoubeirou A Mayaki, Michel Riveill
Long Short Term Memory Fraud Detection Medicare Fraud

Medicare Fraud

Papers

Unsupervised Machine Learning for Explainable Health Care Fraud Detection

Impact of the composition of feature extraction and class sampling in medicare fraud detection

Multiple Inputs Neural Networks for Medicare fraud Detection