EduMind AI

💡 Inspiration

Every student learns differently. Some understand concepts through simple explanations, while others need practice questions, examples, and continuous feedback. However, traditional learning methods often provide the same material and assessments to every student.

We created EduMind AI to explore how Generative AI can make learning more personalized, interactive, and accessible. Our goal is to help students learn directly from their own study materials instead of relying only on generic explanations.

🎯 What It Does

EduMind AI is an AI-powered personalized learning assistant that uses students' PDF study materials to support learning and self-assessment.

Key features include:

• AI Tutor: Students can ask questions about their uploaded PDF and receive explanations based on the material.
• AI Summarizer: Generates structured summaries to help students revise important concepts.
• AI Quiz Generator: Creates multiple-choice questions from study material, with options for 5, 10, or 15 questions and Easy, Medium, or Hard difficulty.
• Automatic Evaluation: Evaluates quiz answers, calculates scores, and provides correct answers with explanations.
• Learning Progress: Tracks quiz attempts, overall accuracy, best scores, and topic-wise performance during the current session.
• Revision Suggestions: Identifies topics with lower quiz accuracy and suggests areas for further practice.

🛠️ How We Built It

EduMind AI was developed using Python and Streamlit to provide a simple, interactive web application.

We used the Google Gemini API to power the AI Tutor, generate summaries, and create multiple-choice quizzes based on uploaded learning materials.

PyMuPDF is used to extract text from PDF documents, while python-dotenv helps manage API credentials locally.

The application processes the extracted learning material and sends relevant text to the Gemini model to generate explanations and assessments. Quiz answers are evaluated automatically, and performance information is maintained using Streamlit session state.

Technologies Used

• Python
• Streamlit
• Google Gemini API
• PyMuPDF
• python-dotenv

🚧 Challenges We Faced

One of the key challenges was designing prompts that could generate useful, relevant questions based on the student's own study material rather than producing generic questions.

Another challenge was structuring AI-generated quiz responses so that questions, answer options, correct answers, and explanations could be evaluated consistently by the application.

We also needed to make the interface straightforward for students, allowing them to upload materials, ask questions, take quizzes, and review their performance without navigating a complicated system.

📚 What We Learned

Through this project, we explored how Generative AI APIs can be integrated into practical educational applications. We gained experience with PDF text extraction, prompt engineering, structured AI responses, Python application development, and session-based progress tracking.

We also learned that AI-powered learning tools need to provide understandable explanations and meaningful feedback, not just generate answers.

🚀 What's Next

Our future plans for EduMind AI include:

• Adding support for images, diagrams, and scanned documents using OCR and multimodal AI.
• Introducing more advanced adaptive assessments based on student performance.
• Saving learning progress across sessions using a database.
• Adding visual learning analytics and personalized study plans.
• Supporting additional learning formats such as presentations and video content.

Our vision is to evolve EduMind AI into a more comprehensive personalized learning platform that adapts to students' individual learning needs.

Built With

• aiartificial
• engineeringmachine
• geminigenerative
• intelligenceeducationpersonalized
• learningai
• learningpymupdfprompt
• pythonstreamlitgoogle
• tutoradaptive