affnanation/insurance-charges-prediction

Model: RiskGuard - Insurance Cost Prediction

This is the model artifact repository for the RiskGuard project, a system for predicting health insurance costs.

The main code repository can be found on GitHub.

Model Description

This model aims to predict the annual medical charges for an insurance customer based on several personal attributes.

• Model Type: Linear Regression
• Key Features: age, bmi, smoker, region, and a risk segment derived from clustering.

Files in this Repository

• prediction_pipeline.joblib: The complete pipeline from preprocessing to the regression model.
• kmeans_model.joblib: The K-Means model used for risk segmentation.
• clustering_scaler.joblib: The StandardScaler used before the clustering process.

How to Use

An example of how to load and use the prediction pipeline with joblib in Python:

import joblib
import pandas as pd

# Load the pipeline
pipeline = joblib.load("prediction_pipeline.joblib")

# Create new sample data
new_data = pd.DataFrame({
    'age': [30],
    'sex': ['male'],
    'bmi': [25.0],
    'children': [2],
    'smoker': ['no'],
    'region': ['southwest']
})

# Make a prediction
predicted_cost = pipeline.predict(new_data)
print(f"Predicted Annual Cost: ${predicted_cost[0]:.2f}")