AI-Based Online Exam System

An AI-driven online exam platform built with Java (Swing) and Python (backend).
The system dynamically generates questions using Google Gemini AI, stores them in JSON format, and provides a Java Swing GUI for students to take the exam.

---

✅ Features

• AI Question Generation: Uses Google Gemini to create unique MCQs.
• JSON Question Storage: Questions saved in data/questions_dataset.json.
• Java Swing GUI: User-friendly interface for taking exams.
• Scoring System: Automatically checks answers and displays the final score.
• Expandable: Future integration with proctoring and voice-based answers.

---

🛠 Tech Stack

• Frontend: Java (Swing)
• Backend: Python 3
• AI Model: Google Gemini API
• Data: JSON
• Libraries:
  • Java: Gson (for JSON parsing)
  • Python: google-generativeai

---

📂 Project Structure

AI-Exam-System/
│
├── java-frontend/
│ ├── AIExamSystem.java # Main Java GUI code
│ ├── lib/gson-2.10.jar # JSON parsing library
│
├── python-backend/
│ ├── question_generator.py # Python script for question generation
│ ├── requirements.txt # Python dependencies
│
├── data/
│ ├── questions_dataset.json # Generated question set
│
├── README.md # Project documentation
├── LICENSE
└── .gitignore

---

⚙️ Installation

1. Clone the Repository

git clone https://github.com/EbrahimAR/AIExamSystem.git
cd AI-Exam-System

2. Set Up Python Backend

Install dependencies:

cd python-backend
pip install -r requirements.txt

Create an environment variable for the Gemini API key:

export GEMINI_API_KEY="your_api_key"

Generate questions:

python question_generator.py "Java" "Beginner"

This will save questions in data/questions_dataset.json.

3. Set Up Java Frontend

Open java-frontend/AIExamSystem.java in IntelliJ, Eclipse, or NetBeans.

Add Gson library:

1. Download gson-2.10.jar

2. Place it in java-frontend/lib/

3. Add it to the project classpath.

Compile and run the Java file.

---

▶️ Usage

1. Generate questions using the Python script.

2. Launch the Java GUI application.

3. Answer all questions displayed in the GUI.

4. Submit answers to see your final score.

---

🔗 Future Enhancements

• Facial Recognition Proctoring using OpenCV.

• Speech Recognition for voice-based answers.

• FastAPI Integration for real-time question generation.

• JWT Authentication for secure user management.

---

👨‍💻 Author

Ebrahim Abdul Raoof

LinkedIn

GitHub

---

📜 License

This project is licensed under the MIT License. See LICENSE for details.