Our team was driven by the desire to present information in a visual, easy-to-read way, so that it may be interpreted and user experience may be smoothed.
What it does
We visualize the data comparisons according to various different axes of information.
How we built it
We used matplotlib, pandas, and seaborn to visualize our data and use feature engineering.
Challenges we ran into
One of the challenges we ran into was how to deal with outliers that affected the way the rest of the data was visualized. Another was finding out that the City of Austin's public data removed their latitude and longitude data just yesterday, preventing us from being able to compare locations of dockless rental scooters versus docked bicycles.
Accomplishments that we're proud of
Making visualizations that were pleasing and easy to interpret, and learning how to use certain libraries and applying that knowledge quickly.
What we learned
We learned the difficulty of determining the important analyses of unorganized data, and the struggles of implementing our inspiration for what we wanted to achieve from our data science.
What's next for BCycle Visualizations
Possible future improvements include a count of how many times which kiosks occur as checkouts versus returns, at which times of the day, overlayed onto a geographical map. Knowing which kiosks frequently get close to running out or getting full would allow BCycle to smooth their users' experiences.