[Neuraminidase] Teach2U

Inspiration

As the old adage goes, "Teaching is the best way to learn". During my role as a peer tutor, I have experienced this firsthand; tutees often ask thought-provoking questions which give us the opportunity to see our learnings from a different perspective and find possible gaps in our knowledge. This opportunity is one I would like to share with other students, especially those who are pursuing knowledge outside of their specialisations.

What it does

Teach2U at its simplest level aims to provide questions for the user on their domain-specific content so that they can find links within the concepts, review, and reflect upon their learnings.

How we built it

We started out the opening day with brainstorming and eventually came to an idea all members could agree upon. Initially, we intended to utilise React for the front-end, but realised that it would make linking with the language models more challenging. By recommendation from other participants, we decided to learn Streamlit as it allowed us to quickly put together a user interface.

Challenges we ran into

None of our members are experienced in artificial intelligence or building web applications, so we found it quite challenging to learn and utilise the frameworks under tight time constraints.

Accomplishments that we're proud of

Building using frameworks that are new to us. Stepping out of our comfort zones to bring our ideas to life.

What we learned

Apart from the technical skills and knowledge we have gained through this hackathon, we have also honed our critical thinking and creativity and learnt to persevere through challenges.

What's next for [Neuraminidase] Teach2U

Better user response validation, so that the agent help point out any misconceptions the user may have. The ability to export the chat conversations as flashcards and mindmaps which the users can use to review and refresh their content

Built With

• huggingface
• langchain
• python
• streamlit