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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Accurate real-time traffic forecast is critical for intelligenttransportation systems (ITS) and it serves as the cornerstone of various smartmobility applications. Though this research area is dominated by deep learning,recent studies indicate that the accuracy improvement by developing new modelstructures is becoming marginal. Instead, we envision that the improvement canbe achieved by transferring the "forecasting-related knowledge" across citieswith different data distributions and network topologies. To this end, thispaper aims to propose a novel transferable traffic forecasting framework:Domain Adversarial Spatial-Temporal Network (DASTNet). DASTNet is pre-trainedon multiple source networks and fine-tuned with the target network's trafficdata. Specifically, we leverage the graph representation learning andadversarial domain adaptation techniques to learn the domain-invariant nodeembeddings, which are further incorporated to model the temporal traffic data.To the best of our knowledge, we are the first to employ adversarialmulti-domain adaptation for network-wide traffic forecasting problems. DASTNetconsistently outperforms all state-of-the-art baseline methods on threebenchmark datasets. The trained DASTNet is applied to Hong Kong's new trafficdetectors, and accurate traffic predictions can be delivered immediately(within one day) when the detector is available. Overall, this study suggestsan alternative to enhance the traffic forecasting methods and providespractical implications for cities lacking historical traffic data.

Domain Adversarial Spatial-Temporal Network: A Transferable Framework for Short-term Traffic Forecasting across Cities