Adaptive dynamic graph learning for forecasting urban multimodal flow
Bibliographic record
Abstract
The increasing diversity and integration of transportation modes is changing urban mobility, resulting in complex spatiotemporal urban flow patterns. The current forecasting models, which typically rely on static or manually defined graph structures, are inadequate for capturing the dynamic spatial heterogeneity and complex cross-modal interactions that are present in real urban systems. To address these limitations, this study introduces a multimodal dynamic graph neural network (MM-DyGNN), a novel deep learning model that is designed for urban multimodal flow prediction. MM-DyGNN introduces three key innovations: (i) a time-varying multimodal graph learning module based on Tucker decomposition that adaptively constructs mode- and time-specific diffusion graphs; (ii) a sparse cross-modal interaction module that employs a top-k strategy to capture the most relevant region–mode dependencies; and (iii) an adaptive multitask learning strategy with uncertainty weighting to balance heterogeneous modal objectives. Comprehensive experiments conducted on real-world urban mobility datasets demonstrate that the MM-DyGNN significantly outperforms the baseline models in terms of forecasting accuracy. Ablation studies further validate the effectiveness of each component and demonstrate the ability of the model to interpret dynamic spatiotemporal dependencies and cross-modal interactions. This work provides a methodological foundation for understanding and managing the evolving complexities of urban mobility.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".