Inspiration
We follow a popular software developer streamer who often asks the chat if the stream was entertaining or not. However, this isn't an efficient solution, as not everyone stays around long enough in the stream to voice if they enjoyed the stream or not.
This is where ChatWatch comes in.
What it does
ChatWatch is an simple but intuitive solution which uses sentiment analysis to provide a summary of how users enjoyed the stream.
How we built it
We used Django and Python to build out the backend. The backend holds the routing for our api and database, as well as the logic of our sentiment analysis code. Our frontend was built with React, ChakraUI, and JavaScript.
Challenges we ran into
The main challenges we ran into were connecting the api to the frontend, as well as improving the accuracy of our sentiment analysis code.
Accomplishments that we're proud of
We're quite proud of the accuracy of the sentiment analysis as it can determine how well a stream goes through the general positivity or negativity of the chat.
What we learned
We learned a lot about self-supervised machine learning in order to develop our sentiment analysis code.
Log in or sign up for Devpost to join the conversation.