2021/04/28 by Jordan Cambe, Krittika D’Silva, Cambe, Jordan +7
Business, Management and Accounting · Engineering · Social Sciences · #Consumer Retail Behavior Studies #Digital Marketing and Social Media #FOS: Computer and information sciences #FOS: Physical sciences #Human Mobility and Location-Based Analysis #Machine Learning (cs.LG) #Physics and Society (physics.soc-ph) #Social and Information Networks (cs.SI) #Urban Design and Spatial Analysis
paper · pdf · doi:10.48550/arxiv.2104.13981
openalex publication_date 2021/04/28 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Understanding the impact that a new business has on the local market\necosystem is a challenging task as it is multifaceted in nature. Past work in\nthis space has examined the collaborative or competitive role of homogeneous\nvenue types (i.e. the impact of a new bookstore on existing bookstores).\nHowever, these prior works have been limited in their scope and explanatory\npower. To better measure retail performance in a modern city, a model should\nconsider a number of factors that interact synchronously. This paper is the\nfirst which considers the multifaceted types of interactions that occur in\nurban cities when examining the impact of new businesses. We first present a\nmodeling framework which examines the role of new businesses in their\nrespective local areas. Using a longitudinal dataset from location technology\nplatform Foursquare, we model new venue impact across 26 major cities\nworldwide. Representing cities as connected networks of venues, we quantify\ntheir structure and characterise their dynamics over time. We note a strong\ncommunity structure emerging in these retail networks, an observation that\nhighlights the interplay of cooperative and competitive forces that emerge in\nlocal ecosystems of retail establishments. We next devise a data-driven metric\nthat captures the first-order correlation on the impact of a new venue on\nretailers within its vicinity accounting for both homogeneous and heterogeneous\ninteractions between venue types. Lastly, we build a supervised machine\nlearning model to predict the impact of a given new venue on its local retail\necosystem. Our approach highlights the power of complex network measures in\nbuilding machine learning prediction models. These models have numerous\napplications within the retail sector and can support policymakers, business\nowners, and urban planners in the development of models to characterize and\npredict changes in urban settings.\n