Health Insurance Premium Prediction using XGboost Regressor
Chaparala Jyothsna, K. Srinivas, Bandi Bhargavi, Akuri Eswar Sravanth, Atmuri Trinadh Kumar, J.N.V.R. Swarup Kumar
Abstract
The suggested work's goal is to anticipate a person's insurance costs and to identify patients with health insurance policies and medical information, regardless of whether or not they have any health problems. Several sorts of health insurance must be anticipated for a patient. It is possible to estimate an individual's health insurance costs based on the level of emergency department treatment they receive depending on the type of health insurance they possess. Multi-Linear, Decision Tree, Random Forest, and Gradient Boosting Regression were some of the regression models employed in this study. After comparing the accuracies, it was determined that Gradient Boosting was the most accurate of all the methods, with an accuracy of 87 percent. Finally, using the best model, the Telegram-integrated chatbot is trained with instructions to communicate with the user and estimates the insurance premium.