Classification of Machine and Deep learning Techniques for Financial Fraud Detection of Healthcare Industry

Abstract

The healthcare industry is one of the most visible areas with vast potential for data collection, both in terms of patient health and economics. Large-scale credit card theft occurs in the healthcare industry because of the steady improvement of electronic payment methods, and credit card fraud monitoring has proven to be a difficult monetary burden for many service providers. Therefore, the method for identifying frauds needs constant improvement. Numerous fraud incidents occur on a regular basis, each with the potential to cause significant monetary loss. Credit card numbers and other private information may be stolen in a variety of ways, including through phishing and Trojan horse programmes that act like viruses. Consequently, there has to be effective technology for detecting various forms of financial fraud. In this paper, we apply a number of machine learning and deep learning techniques to the problem of financial fraud detection of healthcare. We use a number of different algorithms, including Naive Bayes (NB), Logistic Regression (LR), K-Nearest Neighbor (KNN), Random Forest (RF), Convolutional Neural Network (CNN), and deep Artificial Neural Network (ANN), to skew the training of other normal and irregular transaction features. Available data are utilised to assess the quality of the model. Accuracy was represented graphically as 96.1%, 94.8%, 95.89%, 97.58%, 92.3%, and 98.53% for different algorithms, depending on whether they used NB, LR, KNN, RF, CNN, or deep ANN. The findings of the comparison clearly demonstrated that the deep ANN algorithm produces superior outcomes when compared to other methods..