Calendar

Computational Social Science Research Colloquium /Colloquium in Computational and Data Sciences

Brant Horio
Director, Data Science at LMI/CSS PhD Student

The Pedagogy of Zombies: A Case Study of Agent-Based Modeling Competitions for Introducing Complexity, Simulation, and its Real-World Applications

Friday, October 26, 3:00 p.m.
Center for Social Complexity, 3rd Floor Research Hall

All are welcome to attend.

Abstract:  Complexity is pervasive in our daily lives and while academic programs exist to explore, interpret, experiment with, and apply these concepts to better understand the mechanics of our social world, the field is yet to be widely recognized in the mainstream consciousness. Are there engaging instructional methods and tools that can leverage a lower barrier to entry and indoctrinate new scholars into the science of complexity? In this Halloween-themed talk, I present a use case of a simulation modeling competition (and its evolution over several years) that provided preprogrammed agent-based models of a zombie apocalypse. Participants were challenged to explore and formalize human agent behaviors that leveraged their environment and other human agent-agent interactions to hide, evade, and otherwise prevent a grisly human extinction. I will describe the successes and challenges of this experience and a selection of the most creative solutions. I then go on to describe how this competition concept was extended to contemporary challenges that highlighted for participants potential real-world use cases that included combating the zika virus, and fisheries enforcement by the US Coast Guard. I hope for this talk to initiate dialog for how we might continue similar efforts to more easily introduce and propagate the complexity perspective.

Computational Social Science Research Colloquium /
Colloquium in Computational and Data Sciences

J. Brent Williams

Founder and CEO
Euclidian Trust

Improved Entity Resolution as a Foundation for Model Precision 

Friday, November 2, 3:00 p.m.
Center for Social Complexity, 3rd Floor Research Hall

All are welcome to attend.

Abstract: Analyzing behavior, identifying and classifying micro-differentiations, and predicting outcomes relies on the establishment of a core foundation of reliable and complete data linking. Whether data about individuals, families, companies, or markets, acquiring data from orthogonal sources results in significant matching challenges. These matching challenges are difficult because attempts to eliminate (or minimize) false positives yields an increase in false negatives. The converse is true also.

This discussion will focus on the business challenges in matching data and the primary and compounded impact on subsequent outcome analysis. Through practical experience, the speaker led the development and first commercialization of novel approach to “referential matching”. This approach leads to a more comprehensive unit data model (patient, customer, company, etc.), which enables greater computational resolution and model accuracy by hyper-accurate linking, disambiguation, and detection of obfuscation. The discussion also covers the impact of enumeration strategies, data obfuscation/hashing, and natural changes in unit data models over time.

Computational Social Science Research Colloquium /
Colloquium in Computational and Data Sciences

William Lamberti, CSI PhD Student
Department of Computational and Data Sciences
George Mason University

Classifying Pill Spies Using Storks 

Friday, November 9, 3:00 p.m.
Center for Social Complexity, 3rd Floor Research Hall

All are welcome to attend.

Abstract:  Simple and intuitive measures of shape are substantial challenges in image analysis and computer vision.  While measures of shape do exist, there are only a few intuitive and mathematically derived measures for other polygons.  In this talk, a measure, which we call shape proportions, for regular polygons and circles are shown.  From these proportions, we find the corresponding encircled image-histograms for classification purposes.  This method of using shape proportions and encircled image-histograms is called SPEIs (which is pronounced as ‘spy’).  An analysis using simulated and actual shape images were compared to ensure its utility.  Future work regarding applying SPEIs to NIH pill data using stratified over-representative k-folds cross-validation (abbreviated as STORKC, which is pronounced as ‘stork’) will be discussed.

Computational Social Science Research Colloquium /Colloquium in Computational and Data Sciences

Sanjay Nayar
CSS PhD Student

Title: Interlocking Directorates Analysis: Evidence from India BSE-100

Friday, November 30, 3:00 p.m.
Center for Social Complexity, 3rd Floor Research Hall

All are welcome to attend.

Abstract:

Interlocked directorates among companies are common across the world and have been studied quite extensively in the Western World. This study focuses on interlocking directorates, also referred to as inter-organizational elite cooptation (Allen, 1974), among the top 100 publicly traded companies on the Bombay Stock Exchange (BSE-100) in India. The time period analyzed is between 2006 and 2010, the years spanning the recent great recession. While De (2012) looked at the performance effects of interlocking directorates within Indian business groups irrespective of their membership in BSE-100, it did not address in the analysis the key players, cliques, etc., the evolution of the interlocking over time, or any comparisons with the United States. This broad exploratory study is the first to look at the BSE-100 interlocking directorates’ network to see how it has or has not been dominated by a select group of individuals, companies or sectors during 2006-2010, along with the companies’ performances in the longer-term, given their position in the network. Some comparisons are also made with the US market using information available in published papers (Everard, 2002). This study also serves a secondary purpose of being an introduction to the interconnections between some of the biggest players in the Indian Economy/Stock Market and thus would also be of interest to those studying business in India.

Computational Social Science Research Colloquium /
Colloquium in Computational and Data Sciences

Andrew Crooks, Assoc. Professor
Department of Computational and Data Sciences
George Mason University

Computational Modelling of Slums: Progress & Challenges

Friday, December  7,  3:00 p.m.
Center for Social Complexity, 3rd Floor Research Hall

All are welcome to attend.

Abstract:

Over the last 50 years the world has increasingly become more urbanized, much of this growth has occurred in less developed countries, which often lack the resources to accommodate such growth. This has led to the growth of slums, which is estimated to be home for other 1 Billion people. The UN-Habitat projects slum population to double by 2030, which would make them home for 2 in 5 people living in cities. In this talk I will introduce slums, discuss their growth, and provide an overview on what progress has been made to studying and modeling them. This will lead to a discussion of a series of key challenges that need to be addressed if we are to tackle slums from a computational perspective.

Computational Social Science Research Colloquium /
Colloquium in Computational and Data Sciences

Keith Waters, PhD
Schar School of Public Policy
George Mason University

Firm Formation and the Regional Allocation of Labor

Friday, February 01, 3:00 p.m.
Center for Social Complexity Suite, 3rd Floor Research Hall

All are welcome to attend.

Abstract:  The distribution of city-sizes within countries tends to follow a Pareto distribution that satisfies Zipf’s law. Geographically, larger cities tend to be located more distant from one another than smaller cities.  Working towards an explanation of these empirical observations, a geographic extension of Axtell’s agent-based model of endogenous firm formation is presented. The model introduces three components into the underlying model: migration costs, an urban productivity premium, and an urban congestion cost.

Computational Social Science Research Colloquium /
Colloquium in Computational and Data Sciences

John Schuler
Economics PhD Student, George Mason University

Nonparametric Estimation of General Equilibrium Price Vectors

Friday, February 08, 3:00 p.m.
Center for Social Complexity Suite, 3rd Floor Research Hall

All are welcome to attend.

Abstract:

Agent-based economic modeling often requires the determination of an initial equilibrium price vector. Calculating this directly requires algorithms of exponential computational complexity. It is known that a partial equilibrium price can be estimated using a median of trades. This paper explores the possibility of a multivariate generalization of this technique using depth functions as well as alternative methods.

Computational Social Science Research Colloquium /
Colloquium in Computational and Data Sciences

Robert Axtell
Professor, Computational Social Science Program
Department of Computational and Data Sciences, College of Science
Department of Economic, College of Humanities and Social Sciences
George Mason University

Lifetime/Survival/Reliability/Duration Analysis for Computational Model

Friday, February 15, 3:00 p.m.
Center for Social Complexity Suite, 3rd Floor Research Hall

All are welcome to attend.

Abstract:

In a variety of computational models, structures arise, evolve, then disappear, perhaps replaced by other, comparable structures. For example, in some economic models firms form from the interactions of agents, operate for some period of time, and then exit. In housing models, households hold mortgages for finite periods of time before replacing them either due to refinancing or moving to a new house. In political (marketing) models the interests of parties (businesses) are aligned with certain segments of voters (consumers) for a period of time, until competition leads to realignment (brand switching). In environmental policy models specific polluting technologies have finite lifetimes and are eventually replaced by cleaner technologies. In disease models people are infected for varying lengths of time based on their health status, policies, etc. Traffic jams and conflicts have finite duration.

In this talk I will review the mathematical and statistical formalisms of lifetime analysis, also known as survival analysis in biostatistics and reliability analysis in engineering, focusing on the concepts most useful for computational models. Specifically while the former field has concerned itself with censored data (e.g., short clinical trials during which not all patient health outcomes can be observed), and the latter has focused on schemes to manage unreliable equipment, in computational modeling we often need to better understand both age and lifetime distributions of objects in our models, typically have large amounts of quasi-exhaustive ‘data,’ normally know some covariates, and usually work in discrete time.

I will go through the inter-relationships between survival, failure rate, and life expectancy functions, using parametric distributions to illustrate the main ideas. Then I will work through an extended example based on data concerning U.S. business firms, focusing on the connections between firm age and lifetime distributions, which ends with somewhat surprising conclusions, due to high failure rates among young firms (high ‘infant mortality’).

Computational Social Science Research Colloquium /
Colloquium in Computational and Data Sciences

Douglas A. Samuelson
D.Sc., President and Chief Scientist, InfoLogix, Inc.
and
Russell R. Vane III
Future Planner, National Risk Management Center
Cybersecurity and Infrastructure Security Agency
US Department of Homeland Security

Garbage Cans, Lymph Nodes and Cybersecurity: Modeling Organizational Effectiveness

Friday, February 22, 3:00 p.m.
Center for Social Complexity Suite, 3rd Floor Research Hall

All are welcome to attend.

Abstract:

We re-examine and extend the well-known “Garbage Can Model” of Cohen, March and Olsen (1972).  They postulated that organizational choice can be well represented by a garbage can into which problems and solutions are thrown randomly.  When, by random mixing, a solution meets a problem, the problem is solved and removed from the venue.  In 2006, Folcik and Orosz presented an agent-based model of a lymph node, into which blood cells bring foreign substances and objects that are then neutralized by specialized immune system cells.  This model led several social scientists, notably Troitzsch (2008), to point out a strong resemblance to the garbage can model, but now adding the recognition that problems require skill sets which some, but not all, solvers possess.  Matching skill sets is critical to effective performance, and providing the right mix of solver skill sets enables the organization to perform effectively and economically.  We suggest ways to apply this approach to integrated man-machine systems intended to enhance information systems security.  One implication is that some approaches currently popular with policy-makers are highly unlikely to work.