Team: Lynn Chou, Franchezca Layog, & Luke Evans
Inspiration
One aspect of the “city life” in Seattle is learning how to navigate through the hustle and bustle to get to one place from another. For many UW students, this means walking a few miles a day and spending more time on the bus than in class. Students are well acquainted with the public transportation system - buses, trains - but those aren’t always reliable when you’re in a time crunch. Another popular alternative form of transportation is ridesharing through apps like Uber and Lyft. Ridesharing service prices are not kept constant like public transport and are impacted by the supply and demand of rides on a particular day. Our team wants to explore other external factors that drive this demand, such as different weather conditions or the distance traveled for a ride so we can better understand why prices are the way they are, and what they could mean.
What It Does
Libraries Pandas, NumPy, & Plotly.Express
process_csv.py
- Opens and processes CSV files and returns a data frame
cab_and_weather.py
- Takes in Uber and Lyft cab ride data that describes features such as cab type, price, destination, distance traveled, etc.
- Takes in additional weather data containing features such as temperature, rain, time of day, humidity, etc.
- Manipulate and filter data to create visualizations that assist in answering research questions
How We Built It
- Clean Data: Process and prepare datasets
- Filter for Relevant Features: Filter data to specific columns
- Manipulate Data: Group and perform calculations
- Plot Visualizations: Stylize and produce graphs and maps
Challenges
Some challenges we ran into were setting up Visual Studio Code (VS Code) as that was our collaborative platform and processing the data in VS Code due to our abundance of data/data frames.
Accomplishments
We were able to exercise the skills we learned over the past ten weeks, but we also implement a new library to create an engaging and interactive visualization.
What We Learned
As a group: We learnedhow to use Visual Studio Code as a collaborative team (editing the same document) and communicate so we could find time to meet and work on the project.
On the project: One thing we learned is that generally as the price rate for a cab ride rises, distance traveled rises. However, the price rates still vary greatly and someone can pay $10 less per mile on a ride than someone else, even if they traveled the same distance. We also learned that differently priced cab rides maintain the same ratio of frequency to each other as temperature changes and that the frequency of rides in general rises as temperature rises. Finally, we learned that Lyft’s large car service is cheaper than Uber’s on average. Also, the carpooling service for Lyft is cheaper than Uber. The carpooling option for Lyft is significantly cheaper than its standard service, but the carpooling option for Uber is about the same price rate as its standard service.
What's Next
- Use local Uber and Lyft data from Seattle
- Collect public transportation data for a comparison analysis
- Track traffic data to see its impact on cab ride rates
- Collect data from hotter climates to analyze its impacts on cab ride rates
- Test/validate results using a smaller dataset to compare visualizations
Message from the Team
Please view the presentation video to find out more about our project! We go into further detail about our research questions, results, key takeaways, and our future direction.
With your UW email, you can view our slide deck in detail here: https://drive.google.com/file/d/1N2LyNjvGXUSf3_1jMojikPKjqVijWF3H/view?usp=sharing
You may also see how Plotly.Express rendered our interactive visualizations here: https://docs.google.com/presentation/d/1GtQ1pH1DJIJ95qZyHgu67LkcYuvoui0eqOTazkHvASo/edit?usp=sharing
Enjoy :)
Log in or sign up for Devpost to join the conversation.