What question did this study set out to answer?

Develop a unified deep learning framework that models spatial, temporal, and semantic relationships simultaneously to improve urban taxi demand forecasting.

April 26, 2018Open Access

Deep Multi-View Spatial-Temporal Network for Taxi Demand Prediction

View Full Paper

Ask AI

Bookmark

Develop a unified deep learning framework that models spatial, temporal, and semantic relationships simultaneously to improve urban taxi demand forecasting.
Formulated the Deep Multi-View Spatial-Temporal Network (DMVST-Net) combining temporal representations via LSTM, local spatial features via CNN, and semantic correlations between regions with similar temporal patterns.
Trained and evaluated the proposed multi-view architecture on large-scale real-world taxi trip request data against baseline time series and deep learning models.
DMVST-Net demonstrated superior predictive performance over state-of-the-art time series and standalone spatial or temporal deep learning architectures.
Simultaneously capturing local spatial contexts and broader semantic similarities significantly improved taxi demand estimation across complex urban regions.

Discussion

No takes yet. Share an insight, caveat, or question.

Implication

Computational study demonstrates enhanced taxi demand prediction accuracy using multi-view deep networks, indicating improved resource allocation for urban transportation.

Cite This Study

Yao et al. (2018) studied this question.

synapsesocial.com/papers/69fb108417f3234012dd498a https://doi.org/10.1609/aaai.v32i1.11836

View Full Paper

Ask AI

Bookmark

Implication

Computational study demonstrates enhanced taxi demand prediction accuracy using multi-view deep networks, indicating improved resource allocation for urban transportation.

Deep Multi-View Spatial-Temporal Network for Taxi Demand Prediction

Key Points

Discussion

Implication

Cite This Study

Deep Multi-View Spatial-Temporal Network for Taxi Demand Prediction

Key Points

Discussion

Implication

Cite This Study

Also Consider

Also Consider