ed finding spatially clustered communities in networks, contagion on hypergraphs, models of web-browsing behavior, and network perspectives on software development in groups.
The variety of topics and breadth of discussion
served as a reminder of the diversity of applications
and approaches in computational social science.
Symposium participants discussed possibilities for
crossover and collaboration, and agreed on the importance of having a forum such as this symposium
for encouraging new ideas.
Samarth Swarup served as the chair of this symposium. Madhav Marathe, Kiran Lakkaraju, Milind
Tambe, and Cynthia Lakon were coorganizers. The
papers of this symposium were published as AAAI
Press Technical Report FS-13-05.
Gully Burns is project leader of the Biomedical Knowledge
Engineering Group, Information Sciences Institute, University of Southern California.
Yolanda Gil is the principal investigator and project leader
of the Interactive Knowledge Capture research group in the
Information Sciences Institute and also a research professor
in the Department of Computer Science, University of
Yan Liu is an assistant professor in the Department of Computer Science, University of Southern California
Natalia Villanueva-Rosales is a research assistant professor
in the Department of Computer Science and Cyber-ShARE
Center of Excellence, University of Texas at El Paso.
Sebastian Risi is an assistant professor in the Center for
Computer Games Research at the University of Copenhagen, Denmark.
Joel Lehman is a postdoctoral researcher in the Department
of Computer Science at the University of Texas at Austin.
Jeff Clune is an assistant professor of computer science at
the University of Wyoming.
Christian Lebiere is a member of the research faculty in the
Department of Psychology at Carnegie Mellon University.
Paul S. Rosenbloom is a professor in the Department of
Computer Science at the University of Southern California
and a project leader at the Institute for Creative Technologies.
Frank van Harmelen is a professor in knowledge representation and reasoning at the Vrije Universiteit Amsterdam.
James A. Hendler is a professor for artificial intelligence at
the Rensselaer Polytechnic Institute.
Pascal Hitzler is an associate professor for semantic web
technologies at Wright State University.
Krzysztof Janowicz is an assistant professor for geoinfor-matics at University of California, Santa Barbara.
Samarth Swarup is a research scientist at the Virginia Bioinformatics Institute, Virginia Tech.
74 AI MAGAZINE
Social Networks and Social
Contagion: Web Analytics and
Computational Social Science
Computational social science is rapidly becoming a
topic of sustained interest in the computer science
and artificial intelligence communities. These sys-
tems science approaches are drawing in researchers
from domains such as public health, social science,
marketing and advertising, and others. In these do-
mains, the ubiquity of networks and contagionlike
processes on these networks is leading to advances in
network science, epidemiological modeling, and sim-
ulation science. These advances are facilitated by the
availability of vast quantities of social media data.
This symposium brought together researchers
from multiple disciplines, including public health,
game theory, network science, social science, and
business, with the common goal of developing interdisciplinary solutions to these problems.
One major theme of the symposium was the use of
large-scale, data-driven simulation tools for problems
at the intersection of public health, disaster resilience, and social and political science. These synthetic information models are created by combining
multiple sources of data to create a high-fidelity representation of interactions within a social system,
and are being used for modeling phenomena ranging
from epidemic outbreaks, to social unrest, and disaster response.
The symposium included two invited talks.
William Rand (University of Maryland) presented
work on trust, influence, and urgent diffusion in social media, and Lise Getoor (University of California,
Santa Cruz) spoke about entity-based data science.
Another major theme was the use of game theory
for security in domains where networks and contagionlike phenomena occur, such as crime and transportation networks.
A third theme was the use of social networks and
survey data to model the spread of social behaviors
such as adolescent smoking and drinking behaviors.
This work elucidated the need for taking into account
the coevolutionary nature of peer networks and the
spread of behaviors on these networks, since these
phenomena are driven by both homophily and influence. More generally, as another presented paper
discussed, the interactions between agent-internal
cognitive networks (of beliefs, attitudes) and the
agent-external networks (of communication, influence) can lead to surprising nonlinear phenomena.
The symposium also included six short presentations by student scholarship winners, on topics ranging from studying bullying on social media to modeling hospital-acquired infections.
Other topics that were covered by the talks includ-