Calendar

COMPUTATIONAL SOCIAL SCIENCE FRIDAY SEMINAR

Thomas Pike, PhD Student
Computational Social Science
Department of Computational and Data Sciences
George Mason University

Complex Intelligence Preparation of the Battlefield:
An Effort to Operationalize the Integration of Political Theory to Improve Analysis Across the Intelligence Enterprise 

Friday, April 14, 2017,  3:00 p.m.
Center for Social Complexity Suite
3rd Floor, Research Hall

Fifteen years of conflict have shown severe limitations in the United States’ ability to influence foreign populations in pursuit of national objectives. Intertwined within these challenges is the difficulty the U.S. Intelligence Enterprise has in integrating recent research to more effectively analyze foreign populations and support decision makers. The introduction of a Complex Adaptive Systems based meta-framework for intelligence analysts, supported by an agent based model can reduce the cost of analysts learning and applying new research. As a first attempt we adopt the Army’s current framework Intelligence Preparation of the Battlefield (IPB) and begin to formulate one possible meta-perspective Complex Intelligence Preparation of the Battlefield. Complex IPB has the potential to be a constantly improving model that integrates new and emerging theories from economics, communication, politics, demography, game theory and social network analysis to analyze the emergence and contagion of civil conflict in local populations. Complex IPB can assist in identifying regions of potential instability before escalation. Finally, a quasi-global sensitivity analysis identifies effective and efficient policy levers in the face of limited resources.

Computational Social Science Seminar

OPEN MIC

Friday, April 21, 3:00-4:30 p.m.
Center for Social Complexity Suite
Research Hall, 3^rd Floor

The CSS seminar format for Friday, April 21 will be an “Open Mic” session for CSS PhD students to present their research ideas to their peers prior to starting their projects. Peer feedback in encouraged at this event

Some points for presenters to consider are:

• Do you have a paper that is ‘stalled’ and in need of some help to push it to the finish line?
• Is one of your models producing interesting results but also doing wacky things?
• Do you have exciting results but can’t figure out how to visualize/display them?
• Do you need advice on how to calibrate/estimate your theoretical model with data?

Please visit www.cos.gmu.edu/cds/calendar/to see list of upcoming seminar speakers.

We hope to see you on Friday, April 21.

COMPUTATIONAL SOCIAL SCIENCE FRIDAY SEMINAR

Stephen Davies
Hannah Zontine
Department of Computer Science
University of Mary Washington

The ‘Hidden Trump Model’: Modeling social desirability bias through ABMs

Social  desirability bias is a tendency people have to lie about their opinions if they perceive they will be judged or rejected. We present an Opinion Dynamics model in which agents may not be truthful about their opinions when they interact with their social circle. We model two processes through which agents influence one another: an online anonymous process in which agents can interact with anyone and do not fear social rejection, and a face-to-face process where they interact only with friends and may feel compelled to conform. In a political setting, this would apply to a race in which one of the candidates bears a social stigma and therefore some agents are reluctant to voice support for him or her. The results that these nonlinear and asymmetrical processes will have on the overall electorate are not obvious, and are well-suited to an agent-based study.

We hypothesize that this model will produce a “poll bias” of the kind we saw in the 2016 Presidential election — i.e., a significant difference between the number of agents who say they will vote for a candidate and the number who actually do so on election day. We present an analysis of this “Hidden Trump model” and describe the way in which poll bias depends on the strength of the various interaction processes.

Computational Social Science Friday Seminar

Paul Albert, Independent Scholar

Analyzing and Visualizing Data with Tableau

Friday, May 5, 2017 –  3:00 p.m.
Center for Social Complexity Suite
3rd Floor, Research Hall

Abstract:    In January of 2017, Forbes Magazine listed the top technical job skills showing the highest growth in demand from 2011 to 2015. The number three position, with 1,581% growth, was Tableau, a software solution that helps people see and understand their data.

Tableau offers free licenses for academic research.

In this session, Paul Albert will:

• Provide a hands-on overview of Tableau to show how it can help people do more with their data
• Show examples of Tableau data visualizations relevant to the CSS world
• Discuss how Tableau might be able to alter paradigms for sharing academic findings
• Discuss free resources available to learn more about Tableau

Paul recently retired from Tableau and has started graduate studies with the GMU Art History program. His initial focus is on applying quantitative analysis and social theory to art markets. His secondary focus is on exploring how products like Tableau can be used to support new ways of presenting academic research and findings.

While at Tableau, Paul coordinated and conducted over 60 training events for over 1,100 participants. Paul was also one of four finalists, out of a field of over 200 contestants, in the annual Tableau “Viz Wiz” data visualization contest for 2016.

COMPUTATIONAL SOCIAL SCIENCE FRIDAY SEMINAR

Patrick McLaughlin, Program Director and
Oliver Sherouse, Research Analyst
Program for Economic Research on Regulation
Mercatus Center
George Mason University

QuantGov: An Open-Source Platform for Policy Analytics

Friday, May 12, 2017,  3:00 p.m.
Center for Social Complexity Suite
3rd Floor, Research Hall

The Mercatus Center at George Mason University recently launched QuantGov, an open-source policy analytics platform designed to facilitate policy-relevant research.   QuantGov deploys text analysis and machine-learning algorithms to identify the latent governance indicators buried in policy documents, such as legislation or administrative code. QuantGov grew out of the RegData project, which was designed to capture novel metrics in regulations that would advance our understanding of the United States’ federal regulatory process in ways that were previously infeasible.  QuantGov is the next generation in that project’s evolution.

COMPUTATIONAL SOCIAL SCIENCE FRIDAY SEMINAR

Matthew Oldham, PhD Student
Computational Social Science Program
Department of Computational and Data Sciences
George Mason University

The Quest for Living Beta: Investigating the Implication of Shareholder Networks

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

The behavior of financial markets has, and continues, to frustrate investors and academics. With the advent of new approaches, including complex systems and network analysis, the search for an explanation as to how and why markets behavior as they do has greatly expanded, and moved away from the tradition neoclassical approaches that have been beholden to the Efficient Market Hypothesis.

The complex system approach utilizes a number of a concepts in an attempt to understand stock market returns including; imitation, herding, self-organized co-operativity, and positive feedbacks, with many of these features captured by network analysis. In addition, with the meteoric rises of network science has come the realization that the behavior of a system can vary greatly depending on the network structure (the topology) of a system, thus providing further impetus for the use of network analysis in terms of financial markets.

My presentation will detail my recent research of the US Institutional shareholder networks over the period of 2007-10, a period which includes the beginning of the Global Financial Crisis. The research utilized an extensive dataset provided from the Thomson Reuters 13f database, to undertake a temporal analysis of the networks formed between US institutional investors and the stocks in the S&P 500. The analysis makes use of both projected and bipartite networks and uncovers numerous insights regarding relationships between the market in general, stocks and their shareholders. In addition, I will illustrate how the findings can be used in conjunction with an agent-based model to uncover the workings of the stock market.

COMPUTATIONAL SOCIAL SCIENCE FRIDAY SEMINAR

Andrew Crooks, Associate Professor
Computational Social Science
Department of Computational and Data Sciences
George Mason University

ABM for Simulating Spatial Systems: How are we doing?

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

While great advances in modeling have been made, one of the greatest challenges we face is that of understanding human behavior and how people perceive and behave in physical spaces. Can new sources of data (i.e. “big data”) be used to explore the connections between people and places? 

In this presentation,  I will review the current state of art of modeling geographical systems.  I will  highlight the challenges and opportunities through a series of examples that new data can be used to better understand and simulate how individuals behave within geographical systems.

COMPUTATIONAL SOCIAL SCIENCE FRIDAY SEMINAR

Xiaoyi Yuan, PhD Student
Computational Social Science Program
Department of Computational and Data Sciences
George Mason University

Quantifying the Social Debates of Anti-Vaccination on Twitter

Friday, October 6, 3:00 p.m.
Center for Social Complexity Suite
3rd Floor, Research Hall

Measles is one of the leading causes of death among young children. In many developed countries with high measles, mumps, and rubella (MMR) vaccine coverage, measles outbreaks still happen each year. Social media has been one of the dominant information sources to gain vaccination knowledge and thus has also been the focus of the “anti-vaccine movement”. This talk is about two of my recent research projects on this topic. The first one will be introduced briefly, which is an agent-based model demonstrating how a small amount of online anti-vaccine sentiment could have the power of increasing the probability of measles outbreaks significantly. This research inspired me to investigate details of communicative pattern of “anti-vacciners” by analyzing a large twitter dataset (660892 tweets) after the California Disneyland measles outbreak in 2015. This second research has two main parts: first, in order to identify “anti-vacciners”, I used supervised learning to label each tweet as either positive, neutral, or negative opinion towards vaccination. The linear support vector machine model shows good performance on this dataset with an accuracy score of 72% on test data. Second, Louvain’s method for community detection of the retweet network shows the common pattern of social media communities; i.e., overall fragmented but with a few large communities. By investigating the opinion distribution in big communities, however, I discovered that they are highly overlapped, especially within “anti-vacciners”, meaning that they have more frequent communication within their own opinion group than with others. What’s useful for health communication strategies is to look further into the brokers–those who stand between two or more communities. At the end of the talk, I will address details of analyzing the brokerage as well.

Notice and Invitation
Oral Defense of Doctoral Dissertation
Doctor of Philosophy in Computational Sciences and Informatics
Department of Computational and Data Sciences
College of Science
George Mason University

Daniel Pulido
Bachelor of Science, Boston University, 1998
Master of Science, Worcester Polytechnic Institute, 2003

Self-Similar Spin Images for Point Cloud Matching

Friday, October 13, 2017, 10:00 a.m.
Research Hall, Room 162

All are invited to attend.

Committee
Anthony Stefanidis, Dissertation Director
Estela Blaisten-Barojas
Arie Croitoru
Juan Cebral

The rapid growth of Light Detection And Ranging (Lidar) technologies that collect, process, and disseminate 3D point clouds have allowed for increasingly accurate spatial modeling and analysis of the real world. Lidar sensors can generate massive 3D point clouds of a collection area that provide highly detailed spatial and radiometric information. Simultaneously, the growth of other forms of geospatial data (e.g., crowdsourced Web 2.0 data) have provided researchers with a wealth of freely available data sources that cover a variety of geographic areas. However, combining data from disparate sources requires overcoming numerous technical challenges in order to generate products that mitigate their respective disadvantages and combine their advantages.

Therefore, this dissertation addresses the problem of fusing two point clouds from potentially different sources by considering two specific problems: scale matching and feature matching. To address the problem of feature matching we develop a novel feature descriptor referred to as “Self-Similar Spin Images” which combine the concept of local self-similarity with the descriptive power of Spin Images. To address the problem of scale matching we develop a novel scale detection metric referred to as “Self-Similar Keyscale” which analyzes the self-similarity of two point clouds to identify a characteristic scale to match them. Finally, we develop a novel change detection method as a sample use case of the developed matching techniques.

COMPUTATIONAL SOCIAL SCIENCE FRIDAY SEMINAR

Annetta Burger, PhD Candidate
Computational Social Science Program
Department of  Computational and Data Sciences
George Mason University

Organizing Theories for Disaster Study in Computational Social Science

Friday, October 13, 3:00 p.m.
Center for Social Complexity Suite
3rd Floor, Research Hall

Based on evolving understandings of the complex, interconnected characteristics of disasters and social adaptations we propose to organize disaster theories in the interdisciplinary context of complex adaptive systems (CAS). Within this context disasters can be understood to consist of three sets of interacting systems, the socio-ecological system, the system of collective, social behavior, and the individual actor’s cognitive system. Ongoing dynamic forces within each set periodically build to disrupt events and cause failures in the overall system. In this talk, we will explore the dominant theory and frameworks in disaster studies from the perspective of these sets, demonstrate how they explain disasters, and discuss how Computational Social Science methodologies can support theory-building. By identifying evolving understandings and research questions co-located in disaster theory and CAS we find that the CAS features of heterogeneity, flows, interacting subsystems, emergence from self-organization and bottom-up processes, and adaptation and learning are integral to disaster studies.