AbstractsGeography &GIS

Location spoofing refers to a variety of emerging online geographic practices that allow users to hide their true geographic locations. The proliferation of location spoofing in recent years has stirred debate about the reliability and convenience of crowd-sourced geographic information and the use of location spoofing as an effective countermeasure to protect individual geo-privacy and national security. However, these polarized views will not contribute to a solid understanding of the complexities of this trend. Even today, we lack a robust method for detecting location spoofing and a holistic understanding about its multifaceted implications. Framing the issue from a critical realist perspective using a hybrid methodology, this dissertation aims to develop a Bayesian time geographic approach for detecting this trend in social media and to contribute to our understanding of this complex phenomenon from multiple perspectives. The empirical results indicate that the proposed approach can successfully detect certain types of location spoofing from millions of geo-tagged tweets. Drawing from the empirical results, I further qualitatively examine motivation for spoofing as well as other generative mechanisms of location spoofing, and discuss its potential social implications. Rather than conveniently dismissing the phenomenon of location spoofing, this dissertation calls on the GIScience community to tackle this controversial issue head-on, especially when legal decisions or political policies are reached using data from location-based social media. Only then can we promote more effective and trustworthy geographic practices in the age of big data. Advisors/Committee Members: Daniel, Sui (Advisor).