Detecting latent changes in the spatiotemporal pattern of bike sharing is critical to recognizing the impacts of various exogenous factors and significant events. Although extensive studies have investigated the spatiotemporal pattern dynamics in transportation systems, research that explicitly models the changing regularity and captures its latent changes is very scarce. To fill this research gap, we develop a change-point topic model that incorporates multinomial logit models into multi-dimensional latent Dirichlet allocation to decompose the spatiotemporal pattern into a mixture of activity patterns; meanwhile, our model captures the changing regularity of activity prevalence and its latent changes by integrating Dirichlet multinomial regression and hidden Markov models. We estimate the parameters of our model based on collapsed Gibbs sampling. We conduct numerical experiments using publicly available bike sharing trip records collected in New York. Results show that our model successfully distinguishes several meaningful activity categories, such as commuting. Furthermore, the detected pattern changes offer several insights into the impacts of associated events. Our model also improves the goodness of fit and predictive performance for the spatiotemporal attributes of trips.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Detecting Changes in the Spatiotemporal Pattern of Bike Sharing: A Change-Point Topic Model


    Contributors:
    Chen, Xian (author) / Huang, Kun (author) / Jiang, Hai (author)

    Published in:

    Publication date :

    2022-10-01


    Size :

    3271023 byte




    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    English





    SHARING BIKE AND SHARING BIKE SYSTEM

    YUE WEI | European Patent Office | 2019

    Free access


    Examining spatiotemporal changing patterns of bike-sharing usage during COVID-19 pandemic

    Hu, Songhua / Xiong, Chenfeng / Liu, Zhanqin et al. | Elsevier | 2021