All preferences are stored on your device.
2026
It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.
Large-scale urban flow prediction needs models capturing local diffusion and heterogeneous attraction. Existing GNNs rely on neighborhood aggregation, causing over-smoothing and weak functional modeling. DBHGNN decouples propagation into Laplacian diffusion and POI-derived potential modulation, improving OD flow prediction across four U.S. cities with lower training cost.