🏥 Predicting Student Health Risk

An end-to-end Machine Learning project developed for the Kaggle Playground Series, focused on predicting student health risk using feature engineering, ensemble learning, and deep learning techniques.

This project explores multiple machine learning algorithms and ensemble techniques to predict student health risk from demographic and lifestyle-related features.

The complete workflow includes data preprocessing, feature engineering, model training, hyperparameter tuning, ensemble learning, neural network implementation, and Kaggle submission.

• Predict student health risk accurately.
• Compare the performance of multiple machine learning models.
• Improve performance using ensemble learning.
• Implement a Neural Network using PyTorch.
• Optimize the model for the Kaggle evaluation metric.

Competition: Kaggle Playground Series

Files

train.csv
test.csv

The dataset contains demographic, lifestyle, and health-related attributes used to classify student health risk.

Data Collection
        │
        ▼
Data Cleaning
        │
        ▼
Feature Engineering
        │
        ▼
Handling Missing Values
        │
        ▼
Encoding Categorical Features
        │
        ▼
Feature Scaling
        │
        ▼
Model Training
        │
        ▼
Hyperparameter Tuning
        │
        ▼
Ensemble Learning
        │
        ▼
Neural Network (PyTorch)
        │
        ▼
Kaggle Submission

• Logistic Regression
• Gaussian Naive Bayes
• Decision Tree
• Random Forest
• Extra Trees
• HistGradientBoosting
• XGBoost
• LightGBM
• CatBoost
• Voting Classifier
• Stacking Classifier
• Feed Forward Neural Network (PyTorch)

Note: The Kaggle public leaderboard score may differ slightly from local cross-validation results.

• Python
• Pandas
• NumPy
• Matplotlib
• Scikit-learn
• XGBoost
• LightGBM
• CatBoost
• PyTorch
• Jupyter Notebook

Predicting-Student-Health-Risk/
│
├── data/
│   ├── train.csv
│   └── test.csv
│
├── notebooks/
│   ├── Health_Risk_Prediction_ML.ipynb
│   └── Health_Risk_Prediction_ANN.ipynb
│
├── models/
│
├── results/
│
├── images/
│
├── requirements.txt
├── README.md
└── .gitignore

Clone the repository

git clone https://github.com/mayankptdr/Predicting-Student-Health-Risk.git

Move into the project

cd Predicting-Student-Health-Risk

Install dependencies

pip install -r requirements.txt

Launch Jupyter Notebook

jupyter notebook

• Deploy the model using FastAPI
• Build a Streamlit dashboard
• Improve Neural Network performance
• Experiment with TabNet and FT-Transformer
• Automate the complete ML pipeline

Mayank Patidar

AI & Machine Learning Enthusiast

GitHub: https://github.com/mayankptdr

⭐ If you found this project useful, consider giving it a star.