Inspiration

In the 20th century there was a lot of focus on IQ but in the modern day we have started to notice a shift of emphasis from IQ to a person's EQ as well.

What it does

Our system serves as a virtual counselor to initially ease people into the idea of talking to a mental health professional.

How we built it

We focused on building a retrieval based chat-bot system because it offers simplicity to the users but it also allows us to prototype rapidly to gauge efficient responses from the user. Although generative-generative models offer more responses they are usually very sporadic when they are prompted for such deep conversations. It is not worth risking a wrong answer just for the sake of making the technology as advanced as possible.

Challenges we ran into

Some of the challenges we had was having watson understand and make intelligent discoveries based on the training data. It was very labor intensive to optimize watson for every potential encounter it could have.

Accomplishments that we're proud of

we managed to successfully create an automatic deployment system directly to the slack channel.

What we learned

We learned how to efficiently use the ibm watson assistant platform and how to integrate into existing application like web apps, Facebook messenger or slack.

What's next for SPAR.AI

We believe with the integration of our application into the slack allows people to use our product to alleviate stress often found in either groups of students or

Share this project:
×

Updates