Our inspiration was to code an optimization app for travel throughout the area of our data set. Due to the heavy and dense data, we faced many obstacles and had to pivot to something more realistic. While still focusing on the idea around travel, we decided to pursue an algorithm to predict an individuals change in methods of transportation over a series of GPS coordinates. This allows for further useful data extraction and analysis, as we are able to classify an individual's movements through out the day.
What it does
Our app predicts the mode of transportation given GPS coordinates over a range of time.
How we built it
We built it using a random forest classifier model and we evaluated a series of gps coordinates and clustered common predictions in the model.
Challenges we ran into
Accomplishments that we're proud of
We are proud of our ability to create a project in such a short amount of time. Especially given the scope of the project.
What we learned
We learned how to work with big data sets more effectively.
What's next for Predicting Mode of Transportation Shifts Over Time
This could be a door opening for bigger projects needing accurate predictions in modes of transportation.