Graduate Admission Prediction System

Predicts your chances of getting into grad school using machine learning. Also analyzes your Statement of Purpose and tells you what to improve.

What it does

• Predicts admission probability based on GRE, TOEFL, CGPA, etc.
• Shows which factors help or hurt your chances (SHAP analysis)
• Evaluates your SOP using AI and gives scores on 7 criteria
• Suggests improvements for weak areas
• Auto-fills university ratings from world rankings

Tech used

• Frontend: Streamlit
• Backend: FastAPI
• ML Model: Trained on admission dataset
• Explainability: SHAP
• SOP Analysis: Groq API (LLaMA models)

Setup

Prerequisites

• Python 3.8+
• Groq API key (free at https://console.groq.com)

Installation

1. Clone the repo

git clone https://github.com/saishagoel27/Graduate-Admission-Prediction-System
cd Graduate-Admission-Prediction-System

2. Create virtual environment

python -m venv venv

# On Windows
venv\Scripts\activate

# On Mac/Linux  
source venv/bin/activate

3. Install dependencies

pip install -r requirements.txt

4. Add your Groq API key

Create .streamlit/secrets.toml:

GROQ_API_KEY = "your-api-key-here"

Running the app

Option 1: Using the script (Git Bash/Linux/Mac)

chmod +x run_all.sh
./run_all.sh

Option 2: Manual (Windows/any OS)

Terminal 1 - Backend:

uvicorn backend.main:app --reload

Terminal 2 - Frontend:

streamlit run frontend/app.py

Open http://localhost:8501 in your browser.

How to use

1. Fill in your academic details (GRE, TOEFL, CGPA)
2. Enter university name (optional - auto-fills rating)
3. Paste your Statement of Purpose
4. Click "Predict My Admission Chances"
5. Check the sidebar for detailed analysis:
   • SHAP Analysis: see what's helping/hurting
   • SOP Analysis: get scores on clarity, grammar, etc.
   • Recommendations: actionable tips to improve

Project structure

.
├── backend/
│   ├── main.py              # FastAPI routes
│   ├── utils.py             # SOP scoring logic
│   └── models/              # Trained ML models
├── frontend/
│   └── app.py               # Streamlit UI
├── data/
│   ├── admission_data.csv   # Training data
│   └── UpdatedWorldUniRank23.xlsx
├── notebooks/
│   └── admission.ipynb      # Model training notebook
├── .streamlit/
│   └── secrets.toml         # API keys (don't commit!)
├── requirements.txt
└── run_all.sh

Demo

Project.Walkthrough.mp4

Notes

• The SOP analysis uses Groq's LLaMA models (free tier available)
• Predictions are based on historical data - actual results may vary
• University ratings are from 2023 world rankings

Troubleshooting

Backend won't start:

• Make sure port 8000 isn't in use
• Check if all dependencies installed correctly

SOP analysis fails:

• Verify your Groq API key is correct
• Check your internet connection
• Free tier has rate limits - wait a bit and retry

Frontend can't connect to backend:

• Ensure backend is running on port 8000
• Check firewall settings

License

MIT License - do whatever you want with it

Authors

Built by Anishaa and Saisha for NTCC In-House Practical