English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

Taxi demand prediction has recently attracted increasing research interestdue to its huge potential application in large-scale intelligent transportationsystems. However, most of the previous methods only considered the taxi demandprediction in origin regions, but neglected the modeling of the specificsituation of the destination passengers. We believe it is suboptimal topreallocate the taxi into each region based solely on the taxi origin demand.In this paper, we present a challenging and worth-exploring task, called taxiorigin-destination demand prediction, which aims at predicting the taxi demandbetween all region pairs in a future time interval. Its main challenges comefrom how to effectively capture the diverse contextual information to learn thedemand patterns. We address this problem with a novel ContextualizedSpatial-Temporal Network (CSTN), which consists of three components for themodeling of local spatial context (LSC), temporal evolution context (TEC) andglobal correlation context (GCC) respectively. Firstly, an LSC module utilizestwo convolution neural networks to learn the local spatial dependencies of taxidemand respectively from the origin view and the destination view. Secondly, aTEC module incorporates both the local spatial features of taxi demand and themeteorological information to a Convolutional Long Short-term Memory Network(ConvLSTM) for the analysis of taxi demand evolution. Finally, a GCC module isapplied to model the correlation between all regions by computing a globalcorrelation feature as a weighted sum of all regional features, with theweights being calculated as the similarity between the corresponding regionpairs. Extensive experiments and evaluations on a large-scale dataset welldemonstrate the superiority of our CSTN over other compared methods for taxiorigin-destination demand prediction.

Contextualized Spatial-Temporal Network for Taxi Origin-Destination Demand Prediction