Fact Checker

• 

  Front end of our Streamlit App

Inspiration

We wanted a way to verify information that we see and read.

What it does

Compares statements against a known set of documents.

How we built it

LangChain, Tidb, shared development in Google Colab.

Challenges we ran into

Could not get Streamlit working with the Pem file Web scrapping is hard, documents are in different formats and lots of unnecessary data.

Accomplishments that we're proud of

Being able create a database of truthful reference documents that can be used in multiple locations.

What we learned

Tidb is easy to work with especially with groups. This was the first time for some of the team working with Google Colab and Python

What's next for Fact Checker

Get Streamlit connected to Tidb and create a better front end.

Built With

• chatgpt
• langchain
• python
• tidb