End to End Machine Learning Project with Deployment

Student Performance Indicator

Overview

This project analyzes how various factors such as gender, ethnicity, parental education, lunch type, and test preparation courses influence student test scores. Using a dataset of student performance, we apply machine learning models to predict performance and derive insights.

Webapp Snapshot

Key Features

• Comprehensive data analysis to explore the impact of different variables on students' test scores.
• Application of multiple regression models to predict student performance.
• Evaluation of model effectiveness to identify the best predictor of student success.

Getting Started

1. Clone the repository:
2. Navigate to the project directory and install the required packages:
3. Run the Jupyter notebooks to explore the analysis and modeling process.

Data

The dataset used is from Kaggle, including 1000 rows and 8 columns, focusing on student demographics and their academic scores.

Models Evaluated

• Linear Regression
• Decision Tree
• Random Forest
• K-Neighbors Regressor
• Support Vector Machines
• AdaBoost
• XGBoost
• CatBoost

Conclusion

Insights are drawn from the analysis and the best-performing model is highlighted based on the evaluation metrics.

Technologies Used

• Python
• Pandas
• Numpy
• Matplotlib
• Seaborn
• Scikit-learn

License

This project is licensed under the MIT License - see the LICENSE.md file for details.