In building modern intelligent transportation systems like taxi or ride-sharing apps, the accurate prediction of travel time is irreplaceable. It can not only improve the customers’ experience of traveling, but also help taxi drivers manage their routes and orders in a more efficient way. This problem is challenging mainly due to its large dataset and the complex relationship between the model and features. By using New York City’s taxi record in 2017, this paper investigates several machine learning methods to predict the travel time. Since the number of the original feature is small, we introduce several external features that improve the performance of the models. Finally, we present the evaluation results of the proposed method and ablation study on the design of the proposed model.
-
Notifications
You must be signed in to change notification settings - Fork 1
ceruleangu/NYC-Taxi-Trip-Duration-Prediction
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
About
This project investigates several machine learning methods to predict taxi travel time in NYC.
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published