🎓 EduTwin: LLM-Powered Digital Twin of University Students

🚀 Overview

EduTwin is an AI-powered system that creates a Digital Twin of a student using a dynamic Live Learner Profile (LLP) built from real user data. It leverages Large Language Models (LLMs) to simulate student behavior, diagnose weaknesses, predict performance, and generate personalized learning content.

---

💡 Problem Statement

Most education systems treat students uniformly despite differences in:

• Learning pace
• Background knowledge
• Study habits
• Cognitive preferences

EduTwin addresses this by building a personalized AI twin for each student, enabling adaptive and intelligent learning.

---

🧠 Key Features

🔐 Authentication System

• Secure login/signup
• Password hashing and session handling

🗄️ Database-Driven Profiles

• Stores real student data (no synthetic data)
• Persistent and updatable learner profiles

📊 Live Learner Profile (LLP)

• Dynamic representation of student state
• Includes academic, behavioral, and self-reported data

🔍 Weakness Diagnosis

• Identifies weak topics using LLM reasoning

📘 Personalized Explanations

• Tailored explanations based on student profile

📈 Performance Prediction

• Predicts future performance (High / Medium / Low)

📝 Exam Answer Simulation

• Simulates how a student would answer exam questions

---

🏗️ Project Structure

EduTwin/
├── data/
│   ├── generate_data.py
│   └── raw/
├── core/
│   ├── profile_builder.py
│   └── llp_updater.py
├── twin/
│   ├── prompt_engine.py
│   ├── weakness_diagnoser.py
│   ├── explainer.py
│   ├── predictor.py
│   ├── exam_simulator.py
│   └── twin_engine.py
├── database/
│   ├── db.py
│   └── crud.py
├── auth/
│   └── auth.py
├── ui/
│   ├── app.py
│   └── views/
│       ├── student_view.py
│       ├── teacher_view.py
│       └── profile_form.py
├── setup.py
├── requirements.txt
└── README.md

---

⚙️ Tech Stack

• LLM: Groq API
• Backend: Python
• Frontend/UI: Streamlit
• Database: SQLite
• Auth: Custom authentication with hashed passwords

---

🔄 System Workflow

1. User Authentication

   • Student signs up / logs in

2. Profile Creation

   • Student inputs:

     • Academic scores
     • Study habits
     • Confidence levels

3. Database Storage

   • Data stored and managed via CRUD operations

4. LLP Generation

   • Profile Builder creates structured learner profile

5. LLM Twin Engine

   • Uses LLP for:

     • Weakness detection
     • Explanation generation
     • Prediction
     • Simulation

---

🛠️ Installation & Setup

1️⃣ Clone Repository

git clone https://github.com/suryanshdeo/Edu-Twin.git
cd EduTwin

---

2️⃣ Create Virtual Environment

python -m venv venv

Activate:

• Windows:

venv\Scripts\activate

• Mac/Linux:

source venv/bin/activate

---

3️⃣ Install Dependencies

pip install -r requirements.txt

---

4️⃣ Setup Environment Variables

Create a .env file:

GROQ_API_KEY=your_api_key_here

The app loads this .env automatically at startup.
For cloud deployments, set GROQ_API_KEY in the platform's environment-variable settings.

---

5️⃣ Run the Application

streamlit run ui/app.py

---

👨‍🎓 Usage

Student Flow

• Sign up / Log in

• Fill profile form

• Generate learner profile

• Use:

  • Weakness diagnosis
  • Topic explanation
  • Performance prediction
  • Exam simulation

---

Teacher Flow

• View all students
• Analyze class trends
• Identify weak learners

---

📈 Evaluation Metrics

Capability	Metric
Weakness Diagnosis	Precision / Recall
Prediction	Accuracy / F1
Explanation	Human Rating
Simulation	Behavioral Similarity

---

🔐 Security Practices

• Password hashing (bcrypt)
• Environment variable usage for API keys
• No sensitive data stored in repo

---

🔮 Future Enhancements

• 📅 Temporal tracking of student progress
• 🔁 Counterfactual analysis
• 👥 Student clustering
• 🌐 Deployment (Streamlit Cloud)
• 📊 Advanced analytics dashboard

---