We believe in a personalized experience. After learning about Azure and their API, we saw the potential to use facial recognition for the development of something which can cater content to users. That's why we decided to create MoodMatch. It offers the ability to deliver content catered to your current mood directly and fast.

What it does

Our web app uses machine learning and image recognition to determine the emotions of the user. Based on the data collected, we cater top popular content that matches the current mood of the user.

How we built it

We started by testing Azure and seeing how well it works, then we wrote code in Python to connect with the Azure API get and an emotion as output for an image. We used the emotion to connect with the Reddit API and get a list of the top posts of the week related to the emotion. Then we made a website (front-end and back-end) to connect everything and allow the user to take a picture using their webcam and get interesting content with only one click.

Challenges we ran into

The integration of using Microsoft Azure and passing select data to the web app local server took some time. Also an unexpected challenge was that browsers did not allow images to to be taken from the webcam unless that URL contained https:// so it took time to convert our web page to one.

Accomplishments that we're proud of

We are proud seeing our project start to finish obtaining a functional and clean product at the end. At the beginning of the process we planned exactly what components needed to be developed, as well as how each of those components would connect. We made a clear plan, with realistic expectations and we followed it through.

What we learned

No matter how much you try to simplify a project, complications will still arise. When this came along, teamwork and perseverance were the only thing that pushed us through.

What's next for MoodMatch

Hosting our web app online so that we can share it with the entire online community. Diversifying the content that we provide to offer a more personalized experience. Making our website sexier

Share this project: