Incorporating Graph Attention and Recurrent Architectures for City-Wide Taxi Demand Prediction
Taxi demand prediction is one of the key factors in making online taxi hailing services more successful and more popular. Accurate taxi demand prediction can bring various advantages including, but not limited to, enhancing user experience, increasing taxi utilization, and optimizing traffic efficie...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2019-09-01
|
Series: | ISPRS International Journal of Geo-Information |
Subjects: | |
Online Access: | https://www.mdpi.com/2220-9964/8/9/414 |
id |
doaj-8585f5d5f78b4dd299b3e962774fb783 |
---|---|
record_format |
Article |
spelling |
doaj-8585f5d5f78b4dd299b3e962774fb7832020-11-25T01:11:21ZengMDPI AGISPRS International Journal of Geo-Information2220-99642019-09-018941410.3390/ijgi8090414ijgi8090414Incorporating Graph Attention and Recurrent Architectures for City-Wide Taxi Demand PredictionYing Xu0Dongsheng Li1National Lab for Parallel and Distributed Processing (PDL), School of Computer Science, National University of Defense Technology, Changsha 410073, Hunan, ChinaNational Lab for Parallel and Distributed Processing (PDL), School of Computer Science, National University of Defense Technology, Changsha 410073, Hunan, ChinaTaxi demand prediction is one of the key factors in making online taxi hailing services more successful and more popular. Accurate taxi demand prediction can bring various advantages including, but not limited to, enhancing user experience, increasing taxi utilization, and optimizing traffic efficiency. However, the task is challenging because of complex spatial and temporal dependencies of taxi demand. In addition, relationships between non-adjacent regions are also critical for accurate taxi demand prediction, whereas they are largely ignored by existing approaches. To this end, we propose a novel graph and time-series learning model for city-wide taxi demand prediction in this paper. It has two main building blocks, the first one utilize a graph network with attention mechanism to effectively learn spatial dependencies of taxi demand in a broader perspective of the entire city, and the output at each time interval is then transferred to the second block. In the graph network, the edge is defined by an Origin−Destination relation to capture non-adjacent impacts. The second one uses a neural network which is adept with processing sequence data to capture the temporal correlations of city-wide taxi demand. Using a large, real-world dataset and three metrics, we conduct an extensive experimental study and find that our model outperforms state-of-the-art baselines by 9.3% in terms of the root-mean-square error.https://www.mdpi.com/2220-9964/8/9/414taxi demand predictionOrigin–DestinationgraphattentionGRU |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Ying Xu Dongsheng Li |
spellingShingle |
Ying Xu Dongsheng Li Incorporating Graph Attention and Recurrent Architectures for City-Wide Taxi Demand Prediction ISPRS International Journal of Geo-Information taxi demand prediction Origin–Destination graph attention GRU |
author_facet |
Ying Xu Dongsheng Li |
author_sort |
Ying Xu |
title |
Incorporating Graph Attention and Recurrent Architectures for City-Wide Taxi Demand Prediction |
title_short |
Incorporating Graph Attention and Recurrent Architectures for City-Wide Taxi Demand Prediction |
title_full |
Incorporating Graph Attention and Recurrent Architectures for City-Wide Taxi Demand Prediction |
title_fullStr |
Incorporating Graph Attention and Recurrent Architectures for City-Wide Taxi Demand Prediction |
title_full_unstemmed |
Incorporating Graph Attention and Recurrent Architectures for City-Wide Taxi Demand Prediction |
title_sort |
incorporating graph attention and recurrent architectures for city-wide taxi demand prediction |
publisher |
MDPI AG |
series |
ISPRS International Journal of Geo-Information |
issn |
2220-9964 |
publishDate |
2019-09-01 |
description |
Taxi demand prediction is one of the key factors in making online taxi hailing services more successful and more popular. Accurate taxi demand prediction can bring various advantages including, but not limited to, enhancing user experience, increasing taxi utilization, and optimizing traffic efficiency. However, the task is challenging because of complex spatial and temporal dependencies of taxi demand. In addition, relationships between non-adjacent regions are also critical for accurate taxi demand prediction, whereas they are largely ignored by existing approaches. To this end, we propose a novel graph and time-series learning model for city-wide taxi demand prediction in this paper. It has two main building blocks, the first one utilize a graph network with attention mechanism to effectively learn spatial dependencies of taxi demand in a broader perspective of the entire city, and the output at each time interval is then transferred to the second block. In the graph network, the edge is defined by an Origin−Destination relation to capture non-adjacent impacts. The second one uses a neural network which is adept with processing sequence data to capture the temporal correlations of city-wide taxi demand. Using a large, real-world dataset and three metrics, we conduct an extensive experimental study and find that our model outperforms state-of-the-art baselines by 9.3% in terms of the root-mean-square error. |
topic |
taxi demand prediction Origin–Destination graph attention GRU |
url |
https://www.mdpi.com/2220-9964/8/9/414 |
work_keys_str_mv |
AT yingxu incorporatinggraphattentionandrecurrentarchitecturesforcitywidetaxidemandprediction AT dongshengli incorporatinggraphattentionandrecurrentarchitecturesforcitywidetaxidemandprediction |
_version_ |
1725171559780319232 |