Team73_7G_NTUtor

• 

  Generating a question and answering it

• 

  Feedback from answering the question

• 

  Asking the chatbot about a concept

• 

  How we would like the webapp to look like when deployed

Inspiration

Our journey began as NTU students facing a common hurdle: the scarcity of practice materials for exam preparation. Despite having tutorials and lecture slides, the absence of sufficient practice questions made it difficult to gauge our readiness for exams for some modules. The difficulty in finding professors and tutors to consult due to having too many students and busy schedules also serve as a problem. When issues in concepts in some topics arise where it makes it hard to really understand concepts then and cause uncertainty. This challenge inspired us to create a solution that could benefit students everywhere, leading to the birth of our edutech AI chatbot, which we affectionately call NTUtor.

What it does

NTUtor revolutionizes the way students prepare for exams by generating custom, exam-style questions based on the topics they're studying. It provides immediate, constructive feedback on answers, helping students identify areas for improvement. By filling the gap in available study materials, NTUtor ensures that every student has access to comprehensive, personalized exam preparation. It has a chatbot where students can naturally converse with to learn concepts from their notes that are explained differently and in a more concise or focused manner.

How we built it

We leveraged advanced AI technologies with OpenAI, our understanding of educational needs and streamlit as a beautiful demo tool to develop NTUtor. Our chatbot uses AI Agents for the retrieval of subject materials and generates questions and feedback in a precise JSON format. From the same sources it is able to teach on topics in the module and provide help to students who are unsure of certain theorems or concept. This approach ensures compatibility with digital learning platforms and facilitates a seamless study experience.

Challenges we ran into

One of the major challenges was ensuring that the generated questions and feedback were not only accurate but also varied enough to cover a wide range of topics comprehensively. Additionally, optimizing the AI to understand and adapt to different learning styles and preferences required significant testing and refinement.

Accomplishments that we're proud of

We're proud of creating a tool that addresses a real-world problem we faced as students. Seeing NTUtor evolve from an idea to a functional chatbot that can genuinely aid in exam preparation has been incredibly rewarding. We're also proud of our team's perseverance through the development challenges, leading to a product that can make a difference in students' academic lives.

What we learned

Through this project, we gained a deeper understanding of AI's potential in education and the intricacies of developing an AI-powered educational tool. We learned about the importance of user-centered design and the need to continually test and iterate to meet the users' needs effectively.

What's next for NTUtor

Looking ahead, we plan to expand NTUtor's capabilities to include more subjects and deeper customization options to cater to a wider range of learning styles and preferences. We are able to deploy it as shown in the Figma to allow good user experience to excellent learning opportunities. We're also exploring partnerships with educational institutions to integrate NTUtor into their digital learning environments, making it accessible to more students worldwide of all ages from primary to university to educate at a possibly cheaper price too.

Built With

• openai
• python
• streamlit