Students often waste a lot of time and energy trying to find a study space - at campus, the library, coffee shops, or any other public area. We decided to help students find quickly and conveniently find the best available study spaces so they can spend more time on their studies.
How it works
The Android application displays study space "hotspots" and estimated current usage, crowdsourced from students with the app who have studied in that area. The app demonstrates the approximate heatmap of the students currently studying in nearby locations, and the goal is to display forecasts of the room availability in the next 2 hours to help students make choices of where to study, generated using machine learning.
How we built it
Challenges we ran into
Though we have worked with Android apps before, we are not familiar with building one from scratch ourselves. This caused some delays as we acclimated to the platform and the configuration of necessary permissions.
In addition, learning to use the AWS Machine Learning Forecast tool took far longer than anticipated because it is a newer tool with very little documentation and tooling.
Finally, properly collecting location data in the correct format and exporting to a file that could be imported into our machine learning framework was challenging for us due to Android's convoluted file permission systems.
Accomplishments that we're proud of
We're proud of learning a great deal more about Android application development and learning to use machine learning through AWS Forecast.
What's next for StillSpace
We aim to improve the machine learning features and forecasting, and expand the app's features to continuously collect and analyze data in order to provide the most accurate and updated information for the users.
We also aim to add new features such as displaying the locations of the users' friends studying in that area (with permissions of the users) so that the users are able to easily find their friends nearby.