Insurance Claim Prediction

This project applies machine learning techniques to predict insurance charges based on key features like BMI, smoker status, and age. It also includes K-Means clustering to segment risk groups for insurance pricing optimization.

🔍 Key Features

✔ Predicts Insurance Charges using Random Forest Regression

✔ Performs Clustering (K-Means) on BMI & Charges to segment risk groups

✔ Elbow Method to determine the optimal number of clusters

✔ Tkinter GUI for real-time predictions

Machine Learning Models Used

• RandomForestRegressor & Linear Regression Comparison-> Predicts insurance charges
• Statistical Test: Shapiro Normality,Mann- Whitney U test
• K-Means Clustering -> Segments policyholders into risk groups

Installation

Clone the Repository

git clone https://github.com/Beena-Kurian/Insurance_Claim_Prediction.git

cd Insurance-Claim-Prediction

Usage

• Run Jupyter Notebook
• jupyter notebook
• Open Insurance_Claim_Prediction.ipynb and execute the cells This launches a GUI where users can enter age, BMI, smoker status, and get predictions.

Results & Insights

• Smoking significantly increases medical charges.
• K-Means clustering groups policyholders into different risk categories.
• Combining BMI & Smoking leads to better risk assessment.