Practical guidelines for new taxi drivers in NYC && Machine Learning powered fare estimation tool
A UC Berkeley CE290I Final Project
Taxis (including Uber and Lyft) are inseparable parts of the modern transportation system. As the total population and sizes of cities grow steadily in recent years, demand for taxis increases too. U.S. Bureau of Labor Statistics forecasts that the projected percent change in taxi driver employment from 2018 to 2028 is 20 percent, which is significantly higher than the average growth rate for all occupations, 5 percent [1]. With more and more people coming to big cities searching for taxi driver jobs, it will be valuable to provide basic driving guidelines and profit estimation tools for new taxi drivers, which may help shorten their learning time and maximize the profits of their hard work.
This study uses NYC’s signature yellow cabs as a demonstration. Similar analyses can be performed on other dataset such as UBER to other major cities. To produce reasonable guidelines, both taxi trip data and weather condition data of the city have been analyzed using computation knowledge we have learned throughout the semester. In addition, machine learning techniques are used to develop tools for estimating the driving profit.
-
Data Preprocessing: Data Preparation; Data Aggregation
-
Guidelines: Data Exploration
-
Income estimation tool: Fare Prediction
Data source:
- Top 10 MUST KNOW Places
- Most Profitable Locations for Pickup
- Fare Estimation Tool
| Renjie Wu | Dilu Xu |
|---|---|
 |
 |
| PhD Student, Civil and Environmental Engineering, UC Berkeley | M.S. Student, Civil and Environmental Engineering, UC Berkeley |
Contact [email protected] for any questions