Investigating sense of place of the Las Vegas Strip using online reviews and machine learning approaches

Abstract

Public space is essential to urban public life, activities, and events, and has become a key component of design and regeneration schemes. However, visitors’ experience of public space and the subsequent perceptions and sense of place are often difficult to measure because of its intangible nature and the expensive data collection process. This study collected 20,476 online reviews from 18,387 Tripadvisor users from 2008 to 2019 to understand how visitors experience and react to the public spaces on the Las Vegas Strip. We effectively elicited sense of place on the Las Vegas Strip and quantified the popularity and sentiment of each sense of place facet with Latent Dirichlet Allocation (LDA) and logistic regression machine learning methods. The results specified 30 distinct topics related to the Strip, among which the most important ones were exploring different hotels, night scene, people watching, and walking long distances. The visitor experience explored within this study uncovered multiple facets of sense of place on the Strip and suggested urban design strategies and public space management policies related to the programmatic and physical elements of the Strip sidewalks. The study shows how online reviews can provide strong empirical evidence for visitor experience in built environment projects. This approach can be used by landscape architects, urban planners, and policy makers on post-occupancy evaluation and guide redevelopment efforts to provide a full feedback loop.