Temporal and Spatial Analysis of Mobility Data

Session Submission Type: Panel

Abstract

Every human behavior, including communicative behavior, involves four basic elements (4Ws): Who does What at When and Where. Out of these 4Ws, we have accumulated some knowledge about Who and What with traditional research methods (e.g., survey, experiment, and content analysis). However, due to the unavailability of time-stamped and geo-tagged data, little is known about the When (i.e., temporal) and Where (i.e., spatial) elements in human behavior. The increasing popularity of social media, the expanding capabilities of mobile devices, and the rapid advancement of computational methods provide social researchers a “social telescope” (Golder & Macy, 2014, Annual Review of Sociology) to examine the temporal and spatial characteristics of human behavior at different levels of granularities.

The proposed panel focuses on how to model the temporal and spatial features of human behavior with mobility data in a precise and parsimonious way. The interdisciplinary panelists will use empirical examples from their own studies and others’ research to demonstrate: (1) how we can conceptualize and/or operationalize temporal and spatial variables with different mobility datasets (e.g., mobile phone use and mobile apps use); (2) how we can develop empirical/mathematical models to capture and explain the temporal and spatial characteristics of human behavior; and (3) what are the theoretical and/or practical implications in the temporal and spatial analysis of mobility data. Moreover, the panelists will discuss some “critical” questions that have been voiced: What are major challenges communication researchers need to handle in analyzing mobility data? How can communication researchers work with researchers from other disciplines in mobility modeling? How can we deal with personal privacy, replicability, and other legal/ethical dilemmas?

The panel aims to help raise the awareness among communication scholars of opportunities and risks in the modelling of mobility data. Moreover, the panel will try to build an interdisciplinary dialogue on computational research between communication researchers, computer scientists, and research scientists from the industry.

Sub Unit

• Computational Methods

Chair

• Winson Peng, Michigan State U

Participant

• Kenth Engø-Monsen, Telenor

Individual Presentations

• It’s Not Just About Mobility: How Can Mobility Datasets Advance our Understanding of Information Diffusion? - Winson Peng, Michigan State U; Jonathan J. H. Zhu, City U of Hong Kong

• Recommendation With Spatio-Temporal Information - Yizhou Sun, U of California - Los Angeles

• A Multimethod Approach to Collecting and Understanding Mobile Log Data - Jeffrey Boase, U of Toranto

• Analyzing Mobile Phone Data With Network Science - Cheng-Jun Wang, Nanjing U; Xinzhi Zhang, Hong Kong Baptist University