This project focuses on analyzing and predicting insurance costs based on various factors such as age, sex, BMI, number of children, smoking status, and region. The goal is to apply multiple regression models to predict insurance charges and evaluate their performance.
- Programming Language: Python
- Libraries: Pandas, NumPy, Scikit-Learn, Matplotlib, Seaborn
- Tools: Jupyter Notebook
The dataset contains 2771 entries with the following columns:
| Parameter | Description | Content type |
|---|---|---|
| age | Age in years | integer |
| gender | Male or Female | integer (1 or 2) |
| bmi | Body mass index | float |
| no_of_children | Number of children | integer |
| smoker | Whether smoker or not | integer (0 or 1) |
| region | Which US region - NW, NE, SW, SE | integer (1,2,3 or 4 respectively) |
| charges | Annual Insurance charges in USD | float |
Data visualization techniques such as box plots and distribution plots were used to understand the relationships between variables and insurance charges.
The regression plot for charges with respect to bmi.
The box plot for charges with respect to smoker.
Linear Regression was applied to predict insurance charges. The R^2 score obtained for this model was: 0.7504083820289634.
Ridge Regression, a regularized version of Linear Regression, was also applied. The R^2 score obtained for this model was: 0.7115696971487694.
Polynomial Regression was explored to capture non-linear relationships. The R^2 score obtained for this model was: 0.8133405851901191.
The analysis provided insights into the factors affecting insurance costs and demonstrated the effectiveness of different regression models in predicting these costs. Future work could include exploring additional features and advanced modeling techniques.
This project was developed as part of a machine learning study to apply regression techniques to real-world datasets.
To get started with this project, clone the repository and install the necessary dependencies:
git clone https://github.com/burhanahmed1/Insurance-Cost-Analysis.git
cd Insurance-Cost-Analysis
pip install -r requirements.txtOpen the Jupyter notebook:
jupyter notebook Insurance-Cost-Analysis.ipynbContributions are welcome! Please fork this repository and submit pull requests.
This project is licensed under the MIT License.