Skip to main content

About this book

Social computing concerns the study of social behavior and context based on computational systems. Behavioral modeling reproduces the social behavior, and allows for experimenting with and deep understanding of behavior, patterns, and potential outcomes. The pervasive use of computer and Internet technologies provides an unprecedented environment where people can share opinions and experiences, offer suggestions and advice, debate, and even conduct experiments. Social computing facilitates behavioral modeling in model building, analysis, pattern mining, anticipation, and prediction. The proceedings from this interdisciplinary workshop provide a platform for researchers, practitioners, and graduate students from sociology, behavioral and computer science, psychology, cultural study, information systems, and operations research to share results and develop new concepts and methodologies aimed at advancing and deepening our understanding of social and behavioral computing to aid critical decision making.

Table of Contents


Rational Choice Theory: A Forum for Exchange of Ideas between the Hard and Social Sciences in Predictive Behavioral Modeling

The rational choice model of human behavior provides general assumptions that underlie most of the predictive behavioral modeling done in the social sciences. Surprisingly, despite its origins in the work of eminent mathematicians and computer scientists and its current prominence in the social sciences, there has been relatively little interest among hard scientists in incorporating rational choice assumptions into their agent-based analysis of behavior. It is argued here that doing so will introduce greater theoretical generality into agent-based models, and will also provide hard scientists an opportunity to contribute solutions to known weaknesses in the conventional version of rational choice. This in turn will invigorate the dialogue between social scientists and hard scientists studying social phenomena.
Sun-Ki Chai

The SOMA Terror Organization Portal (STOP): social network and analytic tools for the real-time analysis of terror groups

Stochastic Opponent Modeling Agents (SOMA) have been proposed as a paradigm for reasoning about cultural groups, terror groups, and other socioeconomic-political-military organizations worldwide. In this paper, we describe the SOMA Terror Organization Portal (STOP). STOP provides a single point of contact through which analysts may access data about terror groups world wide. In order to analyze this data, SOMA provides three major components: the SOMA Extraction Engine (SEE), the SOMA Adversarial Forecast Engine (SAFE), and the SOMA Analyst NEtwork (SANE) that allows analysts to find other analysts doing similar work, share findings with them, and let consensus findings emerge. This paper describes the STOP framework.
Amy Sliva, V.S. Subrahmanian, Vanina Martinez, Gerardo I. Simari

The Sociology of Alien Rule

Discontent with alien rule is often assumed to be pervasive, if not universal, thus accounting for the absence of an international market in governance services. There is no shortage of explanations of the antipathy to alien rule, and a great deal of corroborative evidence. Many believe that people seem to prefer to be badly ruled by their own kind than better ruled by aliens. Yet if this is true, then identity trumps competence in the assessment of rule, implying that we are all liable to suffer from suboptimal governance. In contrast, this paper argues that the evidence for the pervasiveness of antipathy to alien rule is overdrawn. To that end, it distinguishes between two different types of alien rule -- elected and imposed – provides a brief portrait of each, and suggests that when aliens are confronted with incentives to rule fairly and efficiently, they can gain legitimacy even when they have been imposed.
Michael Hechter

The DoD Encounters the Blogosphere

Social computing is exploding and the imagination of the Department of Defense is overflowing with ways to exploit this brave new world. Do jihadists have Facebooks? Can “we” (the good guys) use our technological genius to discover and surveil “them” (the bad guys) as they express themselves and seek to find an audience in blogs, forums, and other social media? Can “we” find the bad guys? This electronic spygame excites some, while other reject it as unworkable, undoable, and maybe not even desirable. Certainly the question of who “we” are is problematic.
Blogs and related social media provide windows into culture, insights into social processes, opinion and political activity, that much is defensible. In places like Iran, where Westerners have few direct contacts or opportunities to interact with the local people, blogs provide critical understandings of how thousands of people construe the world. The blogs themselves are implicitly biased samples, but if one can account for those biases, one can glean an improved awareness. Conceivably, “adversarial” blogs can enrich our understanding of goals, objectives, and reasoning of those who might support anti-Western violence either implicitly or actively.
The Department of Defense is very aware of blogs. They are consummate bloggers, with hundreds of blogs on the two highly classified networks that serve this community (JWICS and SIPRNet). Like other Westerners, they frequently read commercial and media blogs, such as Slashdot, Slate, Wired, and even Wonkette. They imagine the “adversary” has a mirror image “cyber-underworld,” where there are equivalent facilities. Is there an easy way to find cyber-underworld that is serving as a mass recruitment source? What does this cyber underworld really look like? What do we know about it and what do we need to know to begin to put together the right questions? Is “targeteering” a feasible goal? What sort of research into blogging and social media should the DoD support in order to diminish violent behavior? In order to better understand the social worlds in which the DoD must accomplish their missions? What kinds of boundaries and safeguards should be put around this type of research in order to ensure ethical research behavior? Who should be involved in developing a research agenda for social computing, national security and defense?
Rebecca Goolsby

Integrating Multi-Agent Technology with Cognitive Modeling to Develop an Insurgency Information Framework (IIF)

This research focuses on the application of multi-agent technology and cognitive modeling to develop a decision model for use in analyzing and evaluating behavior strategies of insurgents in Iraq. Insurgency is a complex behavioral process that has many facets. It is a social as well as military problem and a major cause of injury and death for the citizens and military personnel in Iraq. Social computing is concerned with the study of social behavior abstracted by computational models. Computational modeling provides for social system analysis and evaluation. Social computing can be represented as 3-D simulations of real-world events to train, educate or aid in decision making. Users can maneuver through the life-like 3-D Virtual World to observe, record, and measure various behaviors of agents in relation to their environment. The methodology used in modeling these social phenomena is Agent-Oriented Programming (AOP) which is characterized by the use of real-world objects for design and an agent oriented language for implementation. This research integrates the development of multi-agent societies, Electronic Institutions, and Virtual World technology to conduct social computational modeling.
LeeRoy Bronner, Akeila Richards

Stochastic Opponent Modeling Agents: A Case Study with Hezbollah

Stochastic Opponent Modeling Agents (SOMA) have been proposed as a paradigm for reasoning about cultural groups, terror groups, and other socioeconomic- political-military organizations worldwide. In this paper, we describe a case study that shows how SOMA was used to model the behavior of the terrorist organization, Hezbollah. Our team, consisting of a mix of computer scientists, policy experts, and political scientists, were able to understand new facts about Hezbollah of which even seasoned Hezbollah experts may not have been aware. This paper briefly overviews SOMA rules, explains how more than 14,000 SOMA rules for Hezbollah were automatically derived, and then describes a few key findings about Hezbollah, enabled by this framework.
Aaron Mannes, Mary Michael, Amy Pate, Amy Sliva, V.S. Subrahmanian, Jonathan Wilkenfeld

An approach to modeling group behaviors and beliefs in conflict situations

In modern theater military operations, increasing attention is being directed to the coordination of military operations with ongoing social and economic redevelopment and the reformation of a viable political process. Recent experiences in Somalia, Bosnia, Afghanistan and Iraq have resulted in the formulation by the US Army of a specific doctrine, known as DIME (Diplomatic, Informational, Military and Economic) to coordinate all aspects of operations. A key technical requirement for DIME is the evolution of models of population belief and behavioral responses to planned tactical operations. In this paper, we describe an approach to modeling the responses of population groups to overt manipulations over time periods of days to weeks. The model is derived from extensive past successful work by the authors in using computational cognitive models to reason about the beliefs, desires and intentions of individuals and groups.
Norman D. Geddes, Michele L. Atkinson

Computational Models of Multi-National Organizations

An algorithm for designing multi-national organizations that takes into account cultural dimensions is presented and an example from the command and control field is used to illustrate the approach.
A.H. Levis, Smriti K. Kansal, A.E. Olmez, Ashraf M. AbuSharekh

Clustering of Trajectory Data obtained from Soccer Game Records – A First Step to Behavioral Modeling –

Ball movement in a soccer game can be measured as a trajectory on two dimentional plane, which summarizes the tactic or strategy of game players. This paper gives a first step to extract knowledge about strategy of soccer game by using clustering of trajectory data, which consists of the following two steps. First, we apply a pairwise comparison of two trajectories using multiscale matching. Next, we apply rough-set based clustering technique to the similarity matrix obtained by the pairwise comparisons. Experimental results demonstrated that the method could discover some interesting pass patterns that may be associated with successful goals.
Shoji Hirano, Shusaku Tsumoto

Mobile Phone Data for Inferring Social Network Structure

We analyze 330,000 hours of continuous behavioral data logged by the mobile phones of 94 subjects, and compare these observations with self-report relational data. The information from these two data sources is overlapping but distinct, and the accuracy of self-report data is considerably affected by such factors as the recency and salience of particular interactions. We present a new method for precise measurements of large-scale human behavior based on contextualized proximity and communication data alone, and identify characteristic behavioral signatures of relationships that allowed us to accurately predict 95% of the reciprocated friendships in the study. Using these behavioral signatures we can predict, in turn, individual-level outcomes such as job satisfaction.
Nathan Eagle, Alex (Sandy) Pentland, David Lazer

Human Behavioral Modeling Using Fuzzy and Dempster–Shafer Theory

Human behavioral modeling requires an ability to represent and manipulate imprecise cognitive concepts. It also needs to include the uncertainty and unpredictability of human action. We discuss the appropriateness of fuzzy sets for representing human centered cognitive concepts. We describe the technology of fuzzy systems modeling and indicate its the role in human behavioral modeling. We next introduce some ideas from the Dempster-Shafer theory of evidence. We use the Dempster-Shafer theory to provide a machinery for including randomness in the fuzzy systems modeling process. This combined methodology provides a framework with which we can construct models that can include both the complex cognitive concepts and unpredictability needed to model human behavior.
Ronald R. Yager

Online Behavioral Analysis and Modeling Methodology (OBAMM)

This paper introduces a novel method of tracking user computer behavior to create highly granular profiles of usage patterns. These profiles, then, are used to detect deviations in a users’ online behavior, detecting intrusions, malicious insiders, misallocation of resources, and out-of-band business processes. Successful detection of these behaviors significantly reduces the risk of leaking sensitive data, or inadvertently exposing critical assets.
David J. Robinson, Vincent H. Berk, George V. Cybenko

Mining for Social Processes in Intelligence Data Streams

This work introduces a robust method for identifying and tracking clandestinely operating sub-nets in an active social network. The methodology is based on the Process Query System (PQS) previously applied to process mining in various physical contexts. Given a collection of process descriptions encoding personal and/or coordinated behavior of social entities, we parse a network’s transactional stream for instances of active processes and assign process states to events and functional entities based on a projection of the evidence onto the process models. Our goal is not only to define the social network, but also to identify and track the dynamic states of functionally coherent sub-networks.We apply our methodology to a real world security task— mining a collection of simulated HUMINT and SIGINT intelligence data (the Ali Baba simulated intelligence data set)— and demonstrate superior results both in partitioning and contextualizing the social network.
Robert Savell, George Cybenko

An Ant Colony Optimization Approach to Expert Identification in Social Networks

In a social network there may be people who are experts on a subject. Identifying such people and routing queries to such experts is an important problem. While the degree of separation between any node and an expert node may be small, assuming that social networks are small world networks, not all nodes may be willing to route the query because flooding the network with queries may result in the nodes becoming less likely to route queries in the future. Given this constraint and that there may be time constraints it is imperative to have an efficient way to identify experts in a network and route queries to these experts. In this paper we present an Ant Colony Optimization (ACO) based approach for expert identification and query routing in social networks. Also, even after one has identified the experts in the network, there may be new emerging topics for which there are not identifiable experts in the network. For such cases we extend the basic ACO model and introduce the notion of composibility of pheromones, where trails of different pheromones can be combined to for routing purposes.
Muhammad Aurangzeb Ahmad, Jaideep Srivastava

15. Attribution-based Anomaly Detection: Trustworthiness in an Online Community

This paper conceptualizes human trustworthiness1 as a key component for countering insider threats in an online community within the arena of corporate personnel security. Employees with access and authority have the most potential to cause damage to that information, to organizational reputation, or to the operational stability of the organization. The basic mechanisms of detecting changes in the trustworthiness of an individual who holds a key position in an organization resides in the observations of overt behavior – including communications behavior – over time. “Trustworthiness” is defined as the degree of correspondence between communicated intentions and behavioral outcomes that are observed over time [27], [25]. This is the degree to which the correspondence between the target’s words and actions remain reliable, ethical and consistent, and any fluctuation does not exceed observer’s expectations over time [10]. To be able to tell if the employee is trustworthy is thus determined by the subjective perceptions from individuals in his/her social network that have direct business functional connections, and thus the opportunity to repeatedly observe the correspondence between communications and behavior. The ability to correlate data-centric attributions, as observed changes in behavior from human perceptions; as analogous to “sensors” on the network, is the key to countering insider threats.
Shuyuan Mary Ho

Particle Swarm Social Model for Group Social Learning in Adaptive Environment

This report presents a study of integrating particle swarm algorithm, social knowledge adaptation and multi-agent approaches for modeling the social learning of self-organized groups and their collective searching behavior in an adaptive environment. The objective of this research is to apply the particle swarm metaphor as a model of social learning for a dynamic environment. The research provides a platform for understanding and insights into knowledge discovery and strategic search in human self-organized social groups, such as human communities.
Xiaohui Cui, Laura L. Pullum, Jim Treadwell, Robert M. Patton, Thomas E. Potok

Social Network Analysis: Tasks and Tools

Social network analysis can provide great insights into systems composed of interacting objects, and have been successfully applied to various domains. With many different ways to analyze social networks, no single tool currently supports all analysis tasks, but some incorporate more functionality than others. Moreover, the emergence of a new class of social network analysis techniques, link mining, presents a new range of analysis support to provide by the tools. This paper introduces representative social network analysis tasks from traditional, link mining, and visualization aspects, and evaluates a set of tools with diverse general characteristics and social network analysis functionality.
Steven Loscalzo, Lei Yu

Behavioral Entropy of a Cellular Phone User

The increase of advanced service offered by cellular networks draws lots of interest from researchers to study the networks and phone user behavior. With the evolution of Voice over IP, cellular phone usage is expected to increase expo- nentially. In this paper, we analyze the behavior of cellular phone users and identify behavior signatures based on their calling patterns. We quantify and infer the re- lationship of a person’s randomness levels using information entropy based on the location of the user, time of the call, inter-connected time, and duration of the call. We use real-life call logs of 94 mobile phone users collected at MIT by the Real- ity Mining Project group for a period of nine months. We are able to capture the user’s calling behavior on various parameters and interesting relationship between randomness levels in individual’s life and calling pattern using correlation coeffi- cients and factor analysis. This study extends our understanding of cellular phone user behavior and characterizes cellular phone users in forms of randomness level.
Santi Phithakkitnukoon, Husain Husna, Ram Dantu

Community Detection in a Large Real-World Social Network

Identifying meaningful community structure in social networks is a hard problem, and extreme network size or sparseness of the network compound the difficulty of the task.With a proliferation of real-world network datasets there has been an increasing demand for algorithms that work effectively and efficiently. Existing methods are limited by their computational requirements and rely heavily on the network topology, which fails in scale-free networks. Yet, in addition to the network connectivity, many datasets also include attributes of individual nodes, but current methods are unable to incorporate this data. Cognizant of these requirements we propose a simple approach that stirs away from complex algorithms, focusing instead on the edge weights; more specifically, we leverage the node attributes to compute better weights. Our experimental results on a real-world social network show that a simple thresholding method with edge weights based on node attributes is sufficient to identify a very strong community structure.
Karsten Steinhaeuser, Nitesh V. Chawla

Where are the slums? New approaches to urban regeneration

This paper reports about an application of autocorrelation methods in order to produce more detailed analyses for urban regeneration policies and programs. Generally, a municipality proposes an area as suitable for a urban regeneration program considering the edge of neighbourhoods, but it is possible that only a part of a neighbourhood is interested by social degradation phenomena. Furthermore, it is possible that the more deteriorated area belongs to two different adjacent neighbourhoods. Compared to classical statistical analyses, autocorrelation techniques allow to discover where the concentration of several negative social indicators is located. These methods can determine areas with a high priority of intervention in a more detailed way, thus increasing efficiency and effectiveness of investments in urban regeneration programs. In order to verify the possibility to apply these techniques Bari municipality has been chosen for this research since it shows very different social contexts.
Beniamino Murgante, Giuseppe Las Casas, Maria Danese

21. A Composable Discrete-Time Cellular Automaton Formalism

Existing Cellular Automata formalisms do not consider heterogenous composition of models. Simulations that are grounded in a suitable modeling formalism offer unique benefits as compared with those that are developed using an adhoc combination of modeling concepts and implementation techniques. The emerging and extensive use of CA in simulating complex heterogeneous network systems heightens the importance of formal model specification. An extended discrete-time CA modeling formalism is developed in the context of hybrid modeling with support for external interactions.
Gary R Mayer, Hessam S Sarjoughian

22. Designing Group Annotations and Process Visualizations for Role-Based Collaboration

Team collaboration in situations like emergency management often involves sharing and management of knowledge among distributed domain experts. This study extends our previous research on improving common ground building among collaborators with role-based multiple views, and proposes prototypes of annotation and process visualization tools to further enhance common ground building. We illustrate specific needs through two problem scenarios and propose the designed prototypes through storyboarding.
Gregorio Convertino, Anna Wu, Xiaolong (Luke) Zhang, Craig H. Ganoe, Blaine Hoffman, John M. Carroll

Modeling Malaysian Public Opinion by Mining the Malaysian Blogosphere

Recent confrontations between Malaysian bloggers and Malaysian authorities have called attention to the role of blogging in Malaysian political life. Tight control over the Malaysian press combined with a national encouragement of Internet activities has resulted in an embrace of blogging as an outlet for political expression. In this paper we examine the promise and difficulties of monitoring a local blogosphere from the outside, and estimate the size of the Malaysian ‘sopo’ (social/political) blogosphere.
Brian Ulicny

Reading Between the Lines: Human-centred Classification of Communication Patterns and Intentions

Author identification will benefit from the deeper understanding of human’s approach to infer the intention of an author by some kind of text-content analysis, informal speaking by “read between the lines”. This so-called qualitative text analysis aims to derive purpose, context and tone of a communication. By determining characteristics of text that serve as information carrier for communication intents as well as possible interpretations of those information it is assumed to be feasible to develop advanced computer-based methods. So far, however, only few studies on the information carrier and communication intends are performed. In addition, there is no public benchmark dataset available that can support the development and testing of computational methods.
After giving an overview of the current state of art in research and forensic author identification, this paper details a case study on the identification of text-based characteristics and their linkage to specific individuals or target groups by using indicator patterns of human communication and behaviour. Communication patterns determined in this pilot study can be adopted in several other communication contexts and target groups. Moreover, these pattern found will support further research and development of computational methods for qualitative text analysis and the identification of authors.
Daniela Stokar von Neuforn, Katrin Franke

Metagame Strategies of Nation-States, with Application to Cross-Strait Relations

Metagames (Howard, 1968), a class of formal models in game theory, model the behavior of players who predict each other’s conditional strategies recursively. We present a framework for three-player games based on metagame theory. This framework is well-suited to the analysis of nation-states, especially when the analyst wishes to make few assumptions about the level of recursive reasoning and the preference orderings of players.
Alex Chavez, Jun Zhang

Automating Frame Analysis

Frame Analysis has come to play an increasingly stronger role in the study of social movements in Sociology and Political Science. While significant steps have been made in providing a theory of frames and framing, a systematic characterization of the frame concept is still largely lacking and there are no recognized criteria and methods that can be used to identify and marshal frame evidence reliably and in a time and cost effective manner. Consequently, current Frame Analysis work is still too reliant on manual annotation and subjective interpretation. The goal of this paper is to present an approach to the representation, acquisition and analysis of frame evidence which leverages Content Analysis, Information Extraction and Semantic Search methods to automate frame annotation and provide a systematic treatment of Frame Analysis.
Antonio Sanfilippo, Lyndsey Franklin, Stephen Tratz, Gary Danielson, Nicholas Mileson, Roderick Riensche, Liam McGrath

Using Topic Analysis to Compute Identity Group Attributes

Preliminary experiments are described on modeling social group phenomena that appear to address limitations of social network analysis. Attributes that describe groups independently of any specific members are derived from publication data in a field of science. These attributes appear to explain observed phenomena of group effects on individual behavior without the need for the individual to have a network relationship to any member of the group. The implications of theseresults are discussed.
Lashon B. Booker, Gary W. Strong


Additional information