In an era where athletics play a major part of many lives, it is crucial to pay attention to the potential physical dangers at every corner. Hyper-extension, torn muscles, heat illness, concussions, and other injuries could be life-threatening and career-ending. We were inspired to develop Matched to help coaches and administrators plan soccer games at the optimal times to decrease the risk of athletic injury.

What it does

Matched intelligently plans a 5-day set of soccer games by querying the weather data from a user-provided zip code. Afterwards, it suggests a set of dates and times that optimize heat, humidity, and visibility for the athletes to play.

How we built it

We used node.js as a backend, HTML/CSS/JQuery for the front end, AWS ML to develop the machine learning model, and a linear SVG gradient for the data visualization.

Challenges I ran into

We initially ran into the challenge of finding an appropriate data set. However, we were able to overcome that obstacle after a few hours of searching. We spent most of the early morning having trouble integrating our separate parts that we had been working on; we ran into multiple merge conflicts, integration errors, and other obstacles.

Accomplishments that I'm proud of

We are extremely proud of our venture into machine learning and the data visualization that we learned to use on the way.

What's next for Matched

Currently, we can only pull web data once every three hours due to cost limitations. Before releasing Matched, we would provide updated weather data every hour in order to improve condition forecasting. In addition, we would like to expand the app’s functionality for use in other sports like football or lacrosse.

Share this project: