Identifying spatiotemporal urban activities through linguistic signatures. (November 2018)

• Record Type: Journal Article
• Authors: Fu, Cheng; McKenzie, Grant; Frias-Martinez, Vanessa; Stewart, Kathleen
• Is Part Of: Computers, environment and urban systems. Volume 72(2018)
• Page Start: 25
• Page End: 37
• Publication Date: 2018-11
• Subjects: Twitter -- Natural language processing -- Big data -- Human activity modeling -- Urban dynamics; City planning -- Data processing -- Periodicals; Regional planning -- Data processing -- Periodicals; 303.4834
• DOI: 10.1016/j.compenvurbsys.2018.07.003
• Languages: English
• ISSNs: 0198-9715

Abstract

Abstract: Identifying the activities that individuals conduct in a city is key to understanding urban dynamics. It is difficult, however, to identify different human activities on a large scale without incurring significant costs. This study focuses on modeling the spatiotemporal patterns of different activity types within cities by employing user-contributed, geosocial content as a proxy for human activities. In this work, we use linguistic topic modeling to analyze georeferenced twitter data in order to differentiate different activity types. We then examine the spatial and temporal patterns of the derived activity types in three U.S. cities: Baltimore, MD., Washington, D.C., and New York City, NY. The linguistic patterns reflect the spatiotemporal context of the places where the social media content is posted. We further construct a method to link what people post online to the activities conducted within a city. We then use these derived activities to profile the characteristics of neighborhoods in the three cities, and apply the activity signatures to discover similar neighborhoods both within and between the cities. This approach represents a novel activity-based method for assessing similarity between neighborhoods. Graphical abstract: Highlights: Topic modeling is applied to derive detailed activities in three US cities from Twitter text. Spatiotemporal patterns of activities in the cities are modeled. Derived activities are used as signatures to characterize urban neighborhoods. Jensen-Shannon distance and cosine similarity are used for similarity analysis on neighborhoods' activity signatures.