Everyone likes food, but sometimes certain foods are better for you based on your mood. We did a bit of research on foods that can help boost your mood when you might feel down or depressed. Check out our resource here!
What it does
With a given mood, Alexa will tell you about a restaurant within 5 miles of you that you might be interested in based on how you are feeling.
How we built it
We used AWS Comprehend and AWS Lambda to make an Alexa Skill. AWS Comprehend was used to take the given mood and classify it as positive, negative, neutral, or mixed. We then took that category and returned a nearby restaurant that the user might be interested in.
Challenges we ran into
Originally we wanted to use Google's AutoML along with the Google Places API, but we ran into a lot of trouble crossing platforms from Google to Amazon.
Accomplishments that we're proud of
Even though we didn't end up using it in the end, we trained our first model with Google's AutoML Natural Language Processing. Neither of us have ever done anything with Machine Learning, so this was something we were proud of. Additionally, this is also the first time either of us has ever made a functional Alexa skill, so the fact that we had a working skill in the end is certainly something we're happy with.
What we learned
We definitely learned how to adapt when software doesn't work our way, especially under a time constraint.
What's next for fMoods?
We would want to try to figure out a way to get the APIs from both developers to work correctly.