Oral Candidacy - Lei Meng
|Start:||5/22/2014 at 10:40AM|
|End:||5/23/2014 at 5:00PM|
May 23, 2014, 2:00 pm, 315 Stinson Remick
Dr. Aaron Striegel
Dr. Nitesh Chawla Dr. David Hachen Dr. Gregory Madey
"Computational Strategies for Analyzing Dynamic and Multiplex Networks and Their Social Implications"
With the increasing popularity of smartphones and the wide range of applications available on mobile devices, interactions between people have become easier, as they are no longer limited to traditional ways of communication (e.g., face-to-face interaction, phone call, or SMS) but are also possible through modern ways of smartphone communication via social media websites such as Facebook or Twitter. The mobile data about when/where/who of the interactions provide us with various unprecedented perspectives of people's social lives. A good understanding of people's interactions in social networks as well as in any other types of real-world (e.g., technological or biological) networks would enable us to better understand how the system in question functions.
Traditionally, due to limitations of techniques for data collection, social network research has focused on studying static and homogenous networks. However, in the real-world, social relationships are evolving and are typically of various (sometimes complementary) types. Extracting knowledge from newly available both dynamic and multiplex networks is a non-trivial task due to the lack of methods for their analyses, and consequently many challenging questions have emerged both on the computational as well as the application side.
Therefore, my Ph.D dissertation proposal focuses on developing computational strategies for analyzing dynamic and multiplex networks and studying their social implications. It will first discuss two already completed studies: 1) studying the social and personality impact on smartphone usage behaviors, and 2) systematic analysis of dynamic and heterogenous network data. Next, it will propose a new methodology for learning functional knowledge across heterogenous networks with regard to: 1) transfer learning and 2) consensus clustering.