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

Deep learning methods are being increasingly used for urban trafficprediction where spatiotemporal traffic data is aggregated into sequentiallyorganized matrices that are then fed into convolution-based residual neuralnetworks. However, the widely known modifiable areal unit problem within suchaggregation processes can lead to perturbations in the network inputs. Thisissue can significantly destabilize the feature embeddings and the predictions,rendering deep networks much less useful for the experts. This paper approachesthis challenge by leveraging unit visualization techniques that enable theinvestigation of many-to-many relationships between dynamically variedmulti-scalar aggregations of urban traffic data and neural network predictions.Through regular exchanges with a domain expert, we design and develop a visualanalytics solution that integrates 1) a Bivariate Map equipped with an advancedbivariate colormap to simultaneously depict input traffic and prediction errorsacross space, 2) a Morans I Scatterplot that provides local indicators ofspatial association analysis, and 3) a Multi-scale Attribution View thatarranges non-linear dot plots in a tree layout to promote model analysis andcomparison across scales. We evaluate our approach through a series of casestudies involving a real-world dataset of Shenzhen taxi trips, and throughinterviews with domain experts. We observe that geographical scale variationshave important impact on prediction performances, and interactive visualexploration of dynamically varying inputs and outputs benefit experts in thedevelopment of deep traffic prediction models.

Revisiting the Modifiable Areal Unit Problem in Deep Traffic Prediction with Visual Analytics