Discovering Dependence across Traffic Data of Disparate Regions Using Multiscale Generalized Correlation

Qiu, Dehong; Pin, Zhang; Fang, Shaohong; Lang, Xuwen; Feng, Hao

doi:10.1145/3301551.3301559

Cited by 2 publications

(1 citation statement)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The time‐series‐based method has the advantages of mining the hidden multi‐scale characteristic and analyzing the reasons for the spatio‐temporal heterogeneity of traffic volume. The time‐series‐based method, a data‐driven method based on aggregated traffic volume time series [26], however, rarely considers the impact of spatial distribution dynamics on the traffic volume. Thus, the analysis results of such a method may have “scale mixing” phenomenon or even pseudo‐scale characteristic, which leads to the gap between the understanding of the multi‐scale structure of traffic volume and the reality.…”

Section: Introductionmentioning

confidence: 99%

Analyzing the multi‐scale characteristic for online car‐hailing traffic volume with quantum walk

Niu

Qian

et al. 2022

IET Intelligent Trans Sys

View full text Add to dashboard Cite

The multi‐scale characteristic of online car‐hailing traffic volume can reflect the implied distribution pattern, which is crucial for traffic management and even urban planning. Nevertheless, the spatio‐temporal heterogeneity of online car‐hailing traffic volume makes it challenging to analyze its multi‐scale characteristics effectively. Here, a method named multi‐scale characteristic analysis for online car‐hailing traffic volume with quantum walk (MCATV‐QW) is proposed. MCATV‐QW adopts quantum walks to generate multi‐scale probability patterns that online car‐hailing appears at different locations over time. Then stepwise regression is applied to screen the generated multi‐scale probability patterns, to further analyze the multi‐scale characteristic. We validate MCATV‐QW with online car‐hailing traffic volume in the northeast of Chengdu, China. MCATV‐QW not only achieves better simulation performance, but also reveals the distribution pattern that the influence degree of multi‐scale probability patterns weakens from southwest to northeast of study area. MCATV‐QW also reflects the traffic spatial pattern that is dominated by gradual traffic (48%), with both abrupt (26%) and uniform traffic (26%).

show abstract

Section: Introductionmentioning

confidence: 99%