Events Calendar

PhD Defense - Huy Pham

Dissertation Title
Deriving Real-World Social Strength and Social Influence from Spatiotemporal Data


PhD Candidate
Huy Pham

Committee
Cyrus Shahabi (chair), Yan Liu, Daniel O'Leary (outside member)

Time and Place
Thursday, 13 Aug, 1:00pm
PHE 333 Conference Room

Abstract
The ubiquity of mobile devices and the popularity of location-based services have generated rich datasets of people's location information at a very high fidelity. These location datasets can be used for studying various social behaviors, including social connections and social influence. For example, social studies have shown that people, who are seen together frequently at the same places and the same time, are most probably socially related. Similarly, the fact that a person visits a location by following the recommendation of another person who has visited that same location in the past indicates influence that a person exerts on another.
Correspondingly, this thesis focuses on inferring the real-world social connections and social influence by analyzing people's location information, which are useful in a variety of application domains from sales and marketing to social/cultural studies and intelligence analysis. In particular, in the first two studies of this thesis we propose models (GEOSO and EBM) to not only infer social connections, but also to estimate their strengths quantitatively (aka social strength) by analyzing people's co-occurrences in space and time. In the third study, we first define followship to capture the phenomenon of an individual visiting a real-world location (e.g., restaurant) due the influence of another individual who has visited that same location in the past. Subsequently, we coin the term spatial influence as the concept of inferring pair-wise influence from spatiotemporal data by quantifying the amount of followship influence that an individual has on others, and devise the TLFM model for quantifying followship. In all these studies, we examine the impacts of different factors in the location behaviors of people on social strength and influence, including time, locations and coincidences. We conducted extensive experiments with real-world datasets, which demonstrate the effectiveness of the proposed models in quantifying social strength and influence, and their efficiency in working with large data.

Location: Charles Lee Powell Hall (PHE) - 333

Audiences: Everyone Is Invited

Contact: Lizsl De Leon