Student Performance Predictor

Advanced Machine Learning Ensemble for BCA Student Performance Prediction

A full-featured Django web application that predicts student performance categories (Low, Medium, High) using an ensemble of Random Forest, Decision Tree, and Logistic Regression models. Includes user authentication, prediction history, and an admin dashboard.

Developer: Abhishek Maurya

Screenshots

Landing Page

Input Form

Prediction Results

Features

Machine Learning

Combines 3 ML models with weighted soft voting
Real-time visualizations using Chart.js
Understand which factors matter most
Side-by-side comparison of individual models
Personalized suggestions based on predictions

User Management

Signup/Login with enrollment number
Supports BCA, MCA, B.Tech students
Save and track predictions over time
View all saved predictions
Full admin access to manage users and predictions

Technical

Responsive Bootstrap 5 interface with animations
Environment-based configuration, CSRF protection
Gunicorn + WhiteNoise for deployment

Quick Start

Prerequisites

Python 3.8 or higher
pip (Python package manager)

Installation

Clone the repository

git clone https://github.com/abhishek-maurya576/student-performance-predictor.git
cd student-performance-predictor

Create virtual environment

python -m venv venv
venv\Scripts\activate  # Windows
# or
source venv/bin/activate  # Linux/Mac

Install dependencies
```
pip install -r requirements.txt
```

Set up environment variables

copy .env.example .env  # Windows
# or
cp .env.example .env  # Linux/Mac

Generate SECRET_KEY

python -c "from django.core.management.utils import get_random_secret_key; print(get_random_secret_key())"

Copy the output and paste it in .env file.

Navigate to Django project
```
cd student_performance_predictor
```
Apply database migrations
```
python manage.py migrate
```
Create admin user (optional)
```
python manage.py createsuperuser
```
Run the application
```
python manage.py runserver
```
Open in browser
```
http://127.0.0.1:8000/
```
Admin panel: http://127.0.0.1:8000/admin/

Dataset Features

Academic Performance

Age - Student age (18-30 years)
CGPA Semester 1 & 2 - Cumulative GPA (0-10 scale)
T-1 & T-2 Marks - Test scores (0-40 scale)

Study Habits & Engagement

Study Hours/Week - Weekly study time
Attendance % - Class attendance percentage
Parental Support - Binary indicator
Extracurricular Activities - Count (0-4)
Online Resources Used - Binary indicator

Engineered Features

CGPA Average - Mean of semester CGPAs
T Average - Mean of test scores
CGPA Improvement - Semester 2 - Semester 1
Total Engagement - Sum of support indicators

Machine Learning Models

Ensemble Architecture

Input Features
     ↓
┌────────────────────────────────────┐
│  Feature Engineering Pipeline      │
│  - CGPA Average                    │
│  - Test Average                    │
│  - CGPA Improvement                │
│  - Total Engagement                │
└────────────────────────────────────┘
     ↓
┌────────────────────────────────────┐
│  Three Parallel Models             │
│  ┌──────────────────────────────┐  │
│  │ Random Forest (34% weight)   │  │
│  │ n_estimators=100, depth=10   │  │
│  └──────────────────────────────┘  │
│  ┌──────────────────────────────┐  │
│  │ Decision Tree (29% weight)   │  │
│  │ depth=8, entropy criterion   │  │
│  └──────────────────────────────┘  │
│  ┌──────────────────────────────┐  │
│  │ Logistic Reg. (37% weight)   │  │
│  │ max_iter=200, balanced       │  │
│  └──────────────────────────────┘  │
└────────────────────────────────────┘
     ↓
┌────────────────────────────────────┐
│  Weighted Soft Voting              │
│  final_prob = Σ(weight × prob)     │
└────────────────────────────────────┘
     ↓
  Final Prediction
  (Low/Medium/High)

Model Performance

Cross-Validation Accuracy: 80-87.5%
Ensemble Method: Weighted soft voting
Weights: Based on 5-fold stratified CV

Project Structure

Student_Performance_Predictor/
│
├── .env                          # Environment variables (not in git)
├── .env.example                  # Template for environment setup
├── .gitignore                    # Git ignore rules
├── requirements.txt              # Python dependencies
├── README.md                     # This file
├── render.yaml                   # Render deployment config
├── build.sh                      # Build script for deployment
│
├── training/                     # ML training pipeline
│   ├── train_models.py          # Model training script
│   ├── Training_Data.csv        # Original training dataset
│   └── Cleaned_Training_Data.csv # Preprocessed dataset
│
└── student_performance_predictor/  # Django project
    ├── manage.py
    │
    ├── predictor/               # Main Django app
    │   ├── templates/predictor/
    │   │   ├── landing.html     # Public landing page
    │   │   ├── form.html        # Prediction input form
    │   │   ├── result.html      # Results with charts
    │   │   ├── signup.html      # User registration
    │   │   ├── login.html       # User login
    │   │   └── dashboard.html   # User prediction history
    │   ├── static/predictor/
    │   │   └── style.css        # Custom styles
    │   ├── ml_models/           # Trained models
    │   │   ├── rf_model.pkl
    │   │   ├── dt_model.pkl
    │   │   ├── lr_model.pkl
    │   │   ├── scaler.pkl
    │   │   └── ensemble_config.json
    │   ├── tests/               # Unit tests
    │   │   ├── test_forms.py
    │   │   ├── test_models.py
    │   │   ├── test_views.py
    │   │   └── test_ml_pipeline.py
    │   ├── views.py             # View classes (prediction, auth, dashboard)
    │   ├── models.py            # User and Prediction models
    │   ├── forms.py             # Prediction form
    │   ├── auth_forms.py        # Registration & Login forms
    │   ├── admin.py             # Admin configuration
    │   └── urls.py              # URL routing
    │
    └── student_performance_predictor/  # Django settings
        ├── settings.py          # Configuration
        ├── urls.py              # Main URL config
        ├── wsgi.py              # WSGI config
        └── asgi.py              # ASGI config

User Interface

Landing Page

Attractive hero section with call-to-action
Feature highlights and benefits
Quick access to login/signup

Prediction Form

Clean, intuitive input form
Real-time validation feedback
Grouped sections for easy navigation
Responsive design for all devices

Results Page

Performance Prediction - Clear category display with color coding
Model Comparison Table - Side-by-side results from all models
Interactive Charts:
- Ensemble probability distribution (doughnut chart)
- Model prediction comparison (bar chart)
- Feature importance visualization (horizontal bar chart)
Recommendations - Personalized improvement suggestions
Save Prediction - One-click save for authenticated users

User Dashboard

View all saved predictions
Track performance over time
Delete predictions
Admin view for all users (superuser only)

Authentication Pages

Signup - Register with enrollment number, email, course, and semester
Login - Secure login with remember me option

Technology Stack

Backend

Django 5.1.2 - Web framework
scikit-learn 1.3.2 - Machine learning
pandas 2.2.3 - Data manipulation
numpy 1.26.4 - Numerical computing
joblib 1.4.2 - Model serialization
python-decouple 3.8 - Configuration management
gunicorn 23.0.0 - Production WSGI server
whitenoise 6.8.2 - Static file serving

Frontend

Bootstrap 5 - UI framework
Chart.js - Data visualization
Font Awesome - Icons
Custom CSS - Animations and styling

Data Visualization (Training)

matplotlib 3.9.2 - Plotting library
seaborn 0.13.2 - Statistical visualizations

Model Training

To retrain the models with new data:

Prepare your dataset
- Format: CSV with required columns
- Place in training/Training_Data.csv
Run training script
```
cd training
python train_models.py
```

Models will be saved to

student_performance_predictor/predictor/ml_models/

Restart Django server
```
python manage.py runserver
```

Security

✅ Environment-based configuration
✅ SECRET_KEY in .env file (not committed)
✅ Production security settings
✅ CSRF protection enabled
✅ XSS protection enabled
✅ Secure cookies in production

Deployment

Heroku

# Install Heroku CLI, then:
heroku create your-app-name
heroku config:set SECRET_KEY="your-secret-key"
heroku config:set DEBUG=False
heroku config:set ALLOWED_HOSTS="your-app-name.herokuapp.com"
git push heroku main

Railway

Connect your GitHub repository
Set environment variables in dashboard
Deploy automatically on push

Testing

Run all tests:

cd student_performance_predictor
python manage.py test predictor

Run specific test module:

python manage.py test predictor.tests.test_views
python manage.py test predictor.tests.test_models
python manage.py test predictor.tests.test_forms
python manage.py test predictor.tests.test_ml_pipeline

Database Models

User Model

Custom user model with enrollment-based authentication:

enrollment_number - Unique identifier (used for login)
email - User email (unique)
name - Full name
course - BCA, BCA-MCA(DS), MCA, B.Tech, Other
semester - Current semester (3-8)

Prediction Model

Stores saved predictions:

user - Foreign key to User
Input features - age, cgpa1, cgpa2, ct1_marks, ct2_marks, etc.
predicted_performance - Low (0), Medium (1), High (2)
confidence_score - Ensemble confidence percentage
Individual predictions - rf_prediction, dt_prediction, lr_prediction
Timestamps - created_at, updated_at

URL Routes

Route	View	Description
`/`	LandingPageView	Public landing page
`/predict/`	PredictionView	Prediction form (GET/POST)
`/api/predict/`	PredictionAPIView	JSON API for charts
`/signup/`	SignupView	User registration
`/login/`	LoginView	User login
`/logout/`	LogoutView	User logout
`/dashboard/`	DashboardView	View saved predictions
`/save-prediction/`	SavePredictionView	Save prediction to DB
`/delete-prediction/<pk>/`	DeletePredictionView	Delete a prediction

API Documentation

Prediction API Endpoint

POST /api/predict/

Request Body (Form Data):

{
  "age": 21,
  "cgpa1": 7.5,
  "cgpa2": 8.0,
  "ct1_marks": 30,
  "ct2_marks": 35,
  "study_hours_per_week": 15,
  "attendance_percentage": 80,
  "parental_support": 1,
  "extracurricular_activities": 2,
  "online_resources_used": 1
}

Response:

{
  "final_prediction": {
    "value": 2,
    "label": "High",
    "confidence": 85.3
  },
  "model_predictions": {
    "labels": ["Random Forest", "Decision Tree", "Logistic Regression"],
    "predictions": [2, 2, 2],
    "confidences": [78.5, 82.3, 85.1]
  },
  "ensemble_probabilities": {
    "labels": ["Low", "Medium", "High"],
    "probabilities": [0.05, 0.10, 0.85]
  },
  "feature_importance": {...}
}

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

Fork the repository
Create your feature branch (git checkout -b feature/AmazingFeature)
Commit your changes (git commit -m 'Add some AmazingFeature')
Push to the branch (git push origin feature/AmazingFeature)
Open a Pull Request

Known Issues

Small dataset (50 samples) - may affect generalization
Chart.js version compatibility - some features deprecated

Future Enhancements

License

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

Author

Abhishek Maurya

Acknowledgments

scikit-learn community for ML algorithms
Django team for the excellent web framework
Chart.js for beautiful visualizations
Bootstrap for responsive design components

Support

If you have any questions or run into issues, please:

Check the README documentation above
Open an issue on GitHub
Contact the developer

Show Your Support

If you found this project helpful, please give it a ⭐!

Made with ❤️ by Abhishek Maurya

Name		Name	Last commit message	Last commit date
Latest commit History 12 Commits
assets		assets
student_performance_predictor		student_performance_predictor
training		training
.env.example		.env.example
.gitignore		.gitignore
.python-version		.python-version
LICENSE		LICENSE
README.md		README.md
build.sh		build.sh
render.yaml		render.yaml
requirements.txt		requirements.txt
runtime.txt		runtime.txt

Folders and files

Latest commit

History

Repository files navigation

Student Performance Predictor

Screenshots

Landing Page

Input Form

Prediction Results

Features

Machine Learning

User Management

Technical

Quick Start

Prerequisites

Installation

Dataset Features

Academic Performance

Study Habits & Engagement

Engineered Features

Machine Learning Models

Ensemble Architecture

Model Performance

Project Structure

User Interface

Landing Page

Prediction Form

Results Page

User Dashboard

Authentication Pages

Technology Stack

Backend

Frontend

Data Visualization (Training)

Model Training

Security

Deployment

Heroku

Railway

Testing

Database Models

User Model

Prediction Model

URL Routes

API Documentation

Prediction API Endpoint

Contributing

Known Issues

Future Enhancements

License

Author

Acknowledgments

Support

Show Your Support

About

Resources

License

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

Packages