This Medical Insurance Fraud Detection is trained on the the "Medical Insurance" dataset ( https://www.kaggle.com/rohitrox/healthcare-provider-fraud-detection-analysis ) on kaggle. The model is trained using the "Scikit Learn" library in python using algorithms such as Simple Logistic Regression , Random Forest, Decision tress, XBG boos, etc. After EDA ( Exploratory data analysis ) and feature engineering I was able to achiev a training accuracy of 0.99 and testing accuracy of 0.85 using random trees and grid search.