Skip to content

Exploring a real-world dataset for regression analysis.

Notifications You must be signed in to change notification settings

doruq-IT/Medical_Cost_Analysis

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 
 
 

Repository files navigation

doruq-IT

Introduction

Regression analysis is a fundamental part of machine learning, often used to predict a continuous target variable based on one or more input features. This project focuses on using Lazy Prediction, a method that automates the model selection and evaluation process, combined with the GradientBoostingRegressor model to perform regression analysis on a given dataset.

Project Overview

  1. Data Exploration: The dataset is explored to understand its structure, features, and relationships. Statistical summaries, data visualizations, and correlation analysis are performed.

  2. Lazy Prediction: Lazy Prediction is employed to quickly assess the performance of multiple regression models. It automates the model fitting and prediction process, providing initial insights into the viability of different algorithms.

  3. GradientBoostingRegressor: The GradientBoostingRegressor, a powerful ensemble learning technique, is chosen for further analysis. It is trained, tuned, and evaluated to predict the target variable more accurately.

  4. Model Evaluation: The trained model's performance is evaluated using R2, MAE, and MSE metrics. These metrics provide a quantitative assessment of the model's ability to make accurate predictions.

  5. Results and Conclusion: The project concludes by summarizing the insights gained, the effectiveness of the GradientBoostingRegressor model, and the overall findings of the analysis.

Installation

  1. Clone this repository to your local machine.
  2. Install the required dependencies using the following command:
pip install -r requirements.txt
Usage
Open the Jupyter Notebook Regression_Analysis.ipynb to follow the step-by-step analysis process.
Run each cell in the notebook to generate visualizations, perform analysis, and evaluate the model.
Results
The project demonstrates the following key results:

Initial analysis using Lazy Prediction helps identify promising models quickly.
GradientBoostingRegressor outperforms other models in the prediction task.
The trained model achieves a high R2 score, low MAE, and MSE, indicating its effectiveness in making accurate predictions.
Contributing
Contributions to this project are welcome. You can contribute by opening an issue to suggest improvements or by submitting a pull request with enhancements.

License
This project is licensed under the doruq-IT License.

Releases

No releases published

Packages

No packages published