TRAVEL TIME ESTIMATION FOR TAXI TRIPS USING GPS SEQUENCE
Abstract
With many companies providing taxi services on a click, request and dispatch of taxi has become easier and accessible to a wide range of people. Predicting travel time of a given taxi trip plays important role in many decision making problems that are important from both the service time and economics perspective of this business. In this problem, we aim at using machine learning techniques to predict trip travel times based on the characteristics of the trip. We use the taxi trip data provided by Kaggle to train and test the data. We discuss our approach based on KNN regression using GPS data available as part of trip trajectories.
References
A. de Br´ebisson, ´ E. Simon, A. Auvolat, P. Vincent, and Y. Bengio, “Artificial neural networks applied to taxi destination prediction,” arXiv preprint arXiv:1508.00021, 2015.
B. Guillouet, L. Jean-Michel, P. Besse, and R. Franc¸ois, “Destination prediction by trajectory distribution based model,” 2016.
T. Hoch, “An ensemble learning approach for the kaggle taxi travel time prediction challenge,” ECML-PKDD-DCs, 2015.
Y. Jia, E. Shelhamer, J. Donahue, S. Karayev, J. Long, R. Girshick, S. Guadarrama, and T. Darrell, “Caffe: Convolutional archcture for fast feature embedding,” arXiv preprint arXiv:1408.5093, 2014.
Kaggle, “Pkdd15 taxi trip time prediction,” 2015. [Online]. Available: https://www.kaggle.com/c/pkdd-15-taxi-trip-time-prediction-ii
H. T. Lam, E. Diaz-Aviles, A. Pascale, Y. Gkoufas, and B. Chen, “(blue) taxi destination and trip time prediction from partial trajectories,” arXiv preprint arXiv:1509.05257, 2015.
J. Mendes-Moreira and L. Moreira-Matias, “On learning from taxi-gpstraces,” ECML-PKDD-DCs, 2015.
Additional Files
Published
How to Cite
Issue
Section
License
Copyright (c) 2021 International Education and Research Journal (IERJ)
This work is licensed under a Creative Commons Attribution 4.0 International License.
