2012/06/19 by Vassilis Kostakos, Simo Hosio, Tomi Juntunen +3 · 12 citations
Computer Science · Medicine · Physics and Astronomy · Social Sciences · #Archetype #Complex Network Analysis Techniques #Computer science #Data mining #Data science #Data-Driven Disease Surveillance #Geography #Human Mobility and Location-Based Analysis #Information retrieval #Machine learning #Pedestrian #Proxy (statistics) #Roaming #Social media #World Wide Web #cs.HC
paper · pdf · doi:10.1371/journal.pone.0063980
published in PLoS ONE 8(5), e63980 (Public Library of Science) · 4 pages, 1 figure, 1 table
arxiv created 2012/06/19 · openalex publication_date 2013/05/21 · arxiv updated 2013/06/06 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
Can online behaviour be used as a proxy for studying urban mobility? The increasing availability of digital mobility traces has provided new insights into collective human behaviour. Mobility datasets have been shown to be an accurate proxy for daily behaviour and social patterns, and behavioural data from Twitter has been used to predict real world phenomena such as cinema ticket sale volumes, stock prices, and disease outbreaks. In this paper we correlate city-scale urban traffic patterns with online search trends to uncover keywords describing the pedestrian traffic location. By analysing a 3-year mobility dataset we show that our approach, called Location Archetype Keyword Extraction (LAKE), is capable of uncovering semantically relevant keywords for describing a location. Our findings demonstrate an overarching relationship between online and offline collective behaviour, and allow for advancing analysis of community-level behaviour by using online search keywords as a practical behaviour proxy.