Quizzard

Graph

Inspiration

Research papers are hard to read. Very inaccessible to many people, can be seen as daunting and hard to approach with no prior experiences in the field. Especially for students entering academia and being introduced to research. So we decided to make a tool that can help research papers to be more readable and beginner friendly!

What it does

Takes in a pdf file of a research paper / lecture notes / any similar large content of information. Breaks down the content into more digestible chunks represented as a graph (more visually approachable) which connects to summary of each topic, and the user gets the opportunity to ask more questions based on the pdf or answer quiz questions to test their understanding and get relevant feedback for faster learning.

How we built it

We used python flask front-end and claude api back-end to make a web application.

Challenges we ran into

Figuring out how to deal with getting relevant difficulty questions tailored to the user's abilities. Coming up with xml prompts to ensure the output of the claude's messages were in a specific format aligned with our usage.

Accomplishments that we're proud of

Creating a visual representation (graph) which decomposes the complex nature of the input into something more approachable. Integrating claude's api to assist real life issues.

What we learned

How to use claude api, python flask framework, using json files.

What's next for Quizzard

Improving interface, varying the type of questions (allowing multiple choice / diagrams / games / etc), varying type of input (taking in a video/audio instead of pdf), taking education level as input to provide corresponding summaries that can be understood at that level.