Medical Insurance Claim Using Face Id

Abstract

To protect healthcare providers' finances and stop systematic abuse, it is essential to verify the veracity of medical insurance claims. In order to identify irregularities and fraudulent activity in health insurance claims, this study investigates a hybrid machine learning architecture that combines Support Vector Machines (SVM), Decision Trees, and Random Forest classifiers. To refine the input variables, a carefully selected dataset was subjected to extensive preprocessing, which included feature transformation, data normalization, and sophisticated dimensionality reduction techniques. GridSearchCV was used for hyperparameter tuning, which systematically found the best parameter combinations to maximize the prediction performance of each model. Metrics including accuracy, precision, recall, and F1-score were used to assess the effectiveness of the classifiers. The results showed that the modified ensemble models—Random Forest in particular—had better detection skills than more conventional methods.