Mental health is a serious issue in student life that is often overlooked — leading to long-term consequences for many. I’ve always been passionate about spreading awareness around this topic, and to contribute in my own way, I’ve built a machine learning-powered project titled Depression Prediction Model.

Here is the story of how I completed the project :

• Collected the dataset from Kaggle, which is based on student responses from an Indian platform.
• Performed data preprocessing, feature selection, and feature engineering to make the dataset as generalizable as possible for broader use.
• Trained multiple machine learning models and developed an Artificial Neural Network (ANN) — which performed best, achieving 85.3% accuracy.
• Created a fully functional web application using Flask (backend) and Bootstrap (frontend) to allow users to input their data and choose a preferred model for prediction.

The project taught me a lot — I got to learn a new model named Yggdrasil Decision Forest (YDF) Google's modern forest framework, succeeding TensorFlow Decision Forests. Also learned more data preprocessing techniques, better handling of real-world data to fine-tuning model hyperparameters for improved performance.

The app is hosted online for testing purpose with some limitations, but on a local machine, you can tweak the app.py settings and use all the models used during training.