Inspiration
The inspiration we had was that we wanted to give students a simpler way to find courses taught by professors that they had a great experience with previously or discover a new professor based on the teaching style that got them the grade that they wanted.
What it does
It is a chatbot that is interactive and allows the user to ask questions based on the extensive database that was compiled and fed to train the chatbot to answer complex questions that the user has regarding how well a professor teaches.
How we built it
We used Python in the backend and React in the frontend.
Challenges we ran into
The main challenge we ran into is how do we define an abstract schema of csv data to give data driven answers to questions regarding quantitative data.
Accomplishments that we're proud of
The accomplishment that we are most proud of is that we pivoted from using Gemini to OpenAI based database schema.
What we learned
We learned about how to adapt to the challenge that ran into regarding hallucinations that the chatbot experiences as well as the database schema that we used previously that was too rigid for our use case.
What's next for HackAI Chatbot
What is next is that we will continuously train it to ensure that more capabilities are brought into the application so that it can deal with different use cases such as TAs.
Built With
- ai
- openai
- python
Log in or sign up for Devpost to join the conversation.