Spatio-Temporal Analytics for Exploring Human Mobility Patterns and Urban Dynamics in the Mobile Age

Abstract

In this research, we present a spatio-temporal analytical framework including spatiotemporal visualization (STV), space-time kernel density estimation (STKDE), and spatio-temporal-autocorrelation-analysis (STAA), to explore human mobility patterns and intra-urban communication dynamics. Experiments were conducted using largescale detailed records of mobile phone calls in a city. The space-time path, time series graphs, vertical Bézier curves, STKDE, STAA, and related techniques in 3D GIS as well as statistical tests have been suggested for different spatio-temporal analysis tasks. We also investigated several statistical measures that extend the classic spatial association indices for spatio-temporal autocorrelation analysis. The spatial order of weighted matrix was found to have more significant effects than the temporal neighbors on influencing the autocorrelation strength of hourly phone calls.

Spatiotemporal enabled Content-based Image Retrieval

“…Spatiotemporal autocorrelation concept refers to the relationship between some variable observed in each of space-time settings and the association with its neighbors. In a previous work, Gao (2015) proposed three global spatiotemporal autocorrelation indices but didn't describe how to decompose them into local versions. As an initial trial, this work focuses on extending Getis-Ord's G The STG * i quantifies the spatiotemporal concentration of adjacent features associated with the target i, and works as an indicator for measuring local association in space and time simultaneously.…”

Identifying Local Spatiotemporal Autocorrelation Patterns of Taxi Pick-ups and Dropoffs

“…Spatiotemporal autocorrelation concept refers to the relationship between some variable observed in each of space-time settings and the association with its neighbors. In a previous work, Gao (2015) proposed three global spatiotemporal autocorrelation indices but didn't describe how to decompose them into local versions. As an initial trial, this work focuses on extending Getis-Ord's G * (Equation 1) (Getis and Ord 1992) The STG * i quantifies the spatiotemporal concentration of adjacent features associated with the target i, and works as an indicator for measuring local association in space and time simultaneously.…”

Research on social media feeds – A GIScience perspective

“…Focusing on the content of Tweets, Grinberg et al (2013) proposed a method to detect semantic patterns to infer clusters of users' real world activity. Gao (2014) developed a probabilistic approach to make place recommendations based on the users' geo-social circles, as extracted from Foursquare. In another study, the authors estimate spatiotemporal mobility flows from Twitter for the area of greater Los Angeles to infer origin-and destination trips (Gao et al 2014).…”

“…Gao (2014) developed a probabilistic approach to make place recommendations based on the users' geo-social circles, as extracted from Foursquare. In another study, the authors estimate spatiotemporal mobility flows from Twitter for the area of greater Los Angeles to infer origin-and destination trips (Gao et al 2014). Results have shown similar pattern when comparing with community survey data.…”