Skip to content

README.md

Latest commit

 

History

History
57 lines (43 loc) · 1.97 KB

README.md

File metadata and controls

57 lines (43 loc) · 1.97 KB

Insurance Charge Predictor

Welcome to the Insurance Charge Predictor project! This application predicts insurance charges based on user-provided information using various machine learning models.

Key Features

  • Data Preprocessing: Cleaning and preparing data.
  • EDA: Understanding data patterns and relationships.
  • Outlier Handling: Managing outliers to improve model performance.
  • Machine Learning Algorithms: Linear Regression, Decision Tree Regressor, and Random Forest Regressor.
  • Model Persistence: Models saved using Pickle.
  • Deployment: Hosted using Streamlit for easy accessibility.

How It Works

Users input the following details:

  • Age
  • BMI
  • Sex
  • Number of Children
  • Smoking Status
  • Region in India

They can then select a machine learning model (Linear Regressor, Decision Tree Regressor, Random Forest Regressor) to predict the insurance charge.

Installation

  1. Clone the repository: 
    git clone https://github.com/yourusername/insurance-charge-predictor.git
  2. Navigate to the project directory: 
    cd insurance-charge-predictor
  3. Install the required dependencies: 
    pip install -r requirements.txt
  4. Run the Streamlit application: 
    streamlit run app.py

Usage

Open your browser and go to http://localhost:8501. Enter the required details, select the machine learning model, and click "Predict" to get the estimated insurance charge.

Models

  • Linear Regression: Assumes a linear relationship between features and target.
  • Decision Tree Regressor: Splits data into subsets based on feature values.
  • Random Forest Regressor: Builds multiple decision trees and merges their predictions.

Thank you for using the Insurance Charge Predictor! We hope this tool provides valuable insights and predictions for your insurance needs.

URL: https://insurance-charge-predictor-gatfzeqefac8tz4ds3hc6x.streamlit.app/