Notes from today's brainstorming meeting, in no particular order.
- Where do different neighborhoods travel to and from at different times of day?
- What the best and worst times to travel between two parts of the city?
- What factors contribute to exceedingly long or short trip times?
- When should you take a subway vs. a taxi, considering variance in driving times?
- What are the potential benefits of carpooling?
- How does behavior vary across drivers? Are there patterns to the more successful or efficient drivers?
- New vs. experienced drivers
- Day vs. night shifts
- Strategies around when to quit (cf. past academic work)?
- Did the speed limit change to 25mph impact taxi fares and/or trip times?
- Predominant cost and location of trips
- Peak season for taxi performance
- Rush hour traffic
- How efficient is the taxi system overall? How often do cabs have a fare, how many "wasted" trips are taken without passengers? Has this changed over the past few years?
- How does weather affect taxi ridership?
- How do special events, holidays, etc., impact traffic flow in the city?
- What explains variance in tipping?
- Could taxi hotspots in the outer boroughs be used to plan new public transportation routes or improve existing ones?
- Could the city subsidize taxi trips to underserved areas (similar to balancing citibike distribution)?
- Yellow vs. green cab performance
- Original data request
- A series of interesting analyses, with GitHub code
- Some tips on working with the data in R
- "Hunting or waiting? Discovering passenger-finding strategies from a large-scale real-world taxi dataset"
- Problems w/ anonymization
- When to quit / driver elasticities
- Optimal strategies?
- Analysis from Berkeley student
- Where most trips happen (by borough)
- "Average taxi" (or randomly sampled one)
- Todd Schneider's analysis
- Visualizing taxi trips
- The trip data can also be accessed through Google BigQuery (some examples)
- Efficiency of urban taxi systems
- Can reduce 20% to 90% of vacant taxi trips
- Special behavior of taxis around holidays
- Why can't you find a taxi in the rain
- Best and worst times of day
- 5am great, lunch bad
- 538 posts on Uber vs TLC
- Uber data
- Outlier Detection in Large-Scale Taxi Trip Records
- Detecting Traffic Anomalies Using Taxi GPS Data
- Effect of Taxi Information System on Efficiency and Quality of Taxi Services (under paywall)
- Measuring the Efficiency of the Urban Taxi Service System:
- Predicting Taxi Pickups in NYC
- YC Taxi Data Prediction
- Predicting Taxi Tip-Rates in NYC
- Deep yellow and corresponding notes / github repo
- Do we have an id for the medallion or for individual drivers?