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/

Global: Zendy Plus annual

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

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Traffic flow forecasting is a crucial task in urban computing. The challengearises as traffic flows often exhibit intrinsic and latent spatio-temporalcorrelations that cannot be identified by extracting the spatial and temporalpatterns of traffic data separately. We argue that such correlations areuniversal and play a pivotal role in traffic flow. We put forward {spacetimeinterval learning} as a paradigm to explicitly capture these correlationsthrough a unified analysis of both spatial and temporal features. Unlike thestate-of-the-art methods, which are restricted to a particular road network, wemodel the universal spatio-temporal correlations that are transferable fromcities to cities. To this end, we propose a new spacetime interval learningframework that constructs a local-spacetime context of a traffic sensorcomprising the data from its neighbors within close time points. Based on thisidea, we introduce local spacetime neural network (STNN), which employs novelspacetime convolution and attention mechanism to learn the universalspatio-temporal correlations. The proposed STNN captures local trafficpatterns, which does not depend on a specific network structure. As a result, atrained STNN model can be applied on any unseen traffic networks. We evaluatethe proposed STNN on two public real-world traffic datasets and a simulateddataset on dynamic networks. The experiment results show that STNN not onlyimproves prediction accuracy by 4% over state-of-the-art methods, but is alsoeffective in handling the case when the traffic network undergoes dynamicchanges as well as the superior generalization capability.

Space Meets Time: Local Spacetime Neural Network For Traffic Flow Forecasting