Students Stress Prediction System

Overview

The Student Stress Prediction System is a machine learning–powered Flask web application that analyzes 18 psychological, physical, and environmental factors to predict a student’s stress level (Low / Moderate / High) with 86.36% accuracy.

• This project fulfills the requirement of using Flask instead of Streamlit and delivers a professional, production-ready interface.

Features

• Clean, modern, fully responsive Flask web interface
• Professional dark-blue corporate design
• 18 interactive sliders with real-time value display
• Instant prediction with color-coded results
• Pure Flask + HTML/CSS
• Model loaded once at server startup for fast predictions
• Zero external frontend frameworks

Project Structure

Stress-Predictor-Flask/ │ ├── app.py # Main Flask application + prediction logic ├── stress_classifier_model.pkl # Trained CatBoost model (86.36% accuracy) ├── templates/ │ └── index.html # Beautiful frontend UI └── static/ └── style.css # Professional dark-blue styling

Dataset Details

• Source: Student Stress Factors Dataset
• Total Features Used: 18
• Target: Stress Level (0 = Low, 1 = Moderate, 2 = High)

Important Input Features (18)

• Anxiety Level (0–21) * Self-Esteem (0–30)
• Mental Health History (0/1) * Depression (0–27)
• Headache (0–5) * Blood Pressure (1–3)
• Sleep Quality (0–5) * Breathing Problem (0–5)
• Noise Level (0–5) * Living Conditions (0–5)
• Safety (0–5) * Basic Needs (0–5)
• Academic Performance (0–5) * Study Load (0–5)
• Teacher-Student Relationship (0–5)
• Future Career Concerns (0–5)
• Social Support (0–3) * Extracurricular Activities (0–5)

Machine Learning Model

• Algorithm: CatBoost Classifier
• Accuracy: 86.36% (excellent for this dataset)
• Saved using joblib as stress_classifier_model.pkl

Steps Performed

1. Data collection & preprocessing
2. Model training with CatBoost
3. Achieved 86.36% accuracy
4. Model saved using joblib
5. Flask web app development
6. Professional UI with sliders & instant prediction

Flask Web UI

• Two-column responsive layout
• Real-time output next to each slider
• Big centered “Predict” button
• Color-coded alert boxes (Success / Warning / Danger)
• Dark-blue gradient background with white card

Run the Application

• pip install flask joblib scikit-learn pandas catboost python app.py Open browser

Installation & Requirements

• Python 3.8+
• Flask
• joblib
• scikit-learn
• pandas
• catboost