SARA | Devpost

Inspiration

Our initial idea came from Pepper and the Nao bot, which are humanoid robots that also had chat bot functions, in which they are able to have conversations with people. We were also heavily influenced by the old chat bot Eliza.

How we built it

We built SARA using State-of-the-art Natural Language Processing techniques such as bag of words and trained our sentiment analysis on Twitter data because of it's availibility to observe mental condition.

Challenges we ran into

We had some difficulty with programming the bot to identify parts of the input, and converting speech to text and text to speech.

Accomplishments that we're proud of

What we learned

With this project, we were able to learn the basics of machine learning (with the help of the online workshop).

What's next for SARA

The next progress we want to make for CoronaChat is to improve and expand on the type of conversations it can make, and also be able to pick up words that could hint dangerous actions by an individual, such as suicide.

Resources

@incollection{SocherEtAl2013:RNTN, title = {{Parsing With Compositional Vector Grammars}}, author = {Richard Socher and Alex Perelygin and Jean Wu and Jason Chuang and Christopher Manning and Andrew Ng and Christopher Potts}, booktitle = {{EMNLP}}, year = {2013} }