Proceedings of the 2019 International Conference of The Computational Social Science Society of the Americas

This book presents the latest research into CSS methods, uses, and results, as presented at the 2019 annual conference of the CSSSA. This conference was held in Santa Fe, New Mexico, October 24 – 27, 2019, at the Drury Plaza Hotel. What follows is a diverse representation of new results and approaches for using the tools of CSS and agent-based modeling (ABM) for exploring complex phenomena across many different domains. Readers will therefore not only have the results of these specific projects on which to build, but will also gain a greater appreciation for the broad scope of CSS, and have a wealth of case-study examples that can serve as meaningful exemplars for new research projects and activities. The Computational Social Science Society of the Americas (CSSSA) is a professional society that aims to advance the field of CSS in all its areas, from fundamental principles to real-world applications, by holding conferences and workshops, promoting standards of scientific excellence in research and teaching, and publishing novel research findings. Contents Editors and Contributors Social Impression of Faces: From Prediction to Modification 1 Introduction 2 Predicting Social Attributes of Faces 3 Creating a Large-Scale Facial Impression Dataset 4 Validating the Algorithm-Augmented Dataset 5 ModifAE: A Modification Model of Social Impressions 5.1 Architecture 5.2 ModifAE Training Procedure 5.3 How ModifAE Works 5.4 Qualitative Evaluation 5.5 Quantitative Evaluation 5.6 Qualitative Interpretations 6 Discussion References Corruption and the Effects of Influence Within Social Networks: An Agent-Based Model of the ``Lava Jato'' Scandal 1 Introduction 2 Background 3 The Model 3.1 Generation of Networks 3.2 Agents 3.3 Initialization of the Model and Agent Interactions 4 Results 5 Conclusion References Resistance of Communities Against Disinformation 1 Introduction 2 Opinion Dynamics 3 Model of Conspirators 4 Results 5 Discussion References Assessing the Potential of Crowd-Shipping for Food Rescue Logistics Using Agent-Based Modeling 1 Introduction 2 Models of Crowd-Shipping Systems 3 Agent-Based Model 3.1 Model Overview 3.2 Sub-Model 1: Restaurant Agent Decision-Making 3.3 Sub-Model 2: Shelter Assignment 3.4 Sub-Model 3: Crowd-Shipper Agent Decision-Making 3.5 Initialization 4 Experimentation and Results 5 Conclusion References Exogenous Shocks Lead to Increased Responsiveness and Shifts in Sentimental Resilience in Online Discussions 1 Introduction 2 Background 3 Methodology 3.1 Data 3.2 Conversation Dynamics 3.3 Sentiment Transfer 3.4 Transfer Entropy 4 Results 4.1 Conversation Dynamics 4.2 Sentiment Transfer 4.3 Total Transfer Entropy 5 Conclusion References The Cat and Mouse of Getting Around the Law 1 Introduction 2 Results 3 Optimizing Strategies 3.1 Fast and Strong (“All in”) 3.2 Slow and Strong (“Slow and Aggressive”) 3.3 Fast and Weak 3.4 Slow and Weak (“Do Nothing”) 4 Discussion 5 Conclusion References The Degree-Dependent Threshold Model: Towards a Better Understanding of Opinion Dynamics on Online Social Networks 1 Introduction 2 Methods 2.1 Data Set and Twitter Analysis Results 2.2 Generating Networks 2.3 Assigning Thresholds 2.4 Running Simulations 3 Simulation Results 4 Conclusion References Modeling Genocide: An Agent-Based Model of Bystander Motivations and Societal Restraints 1 Introduction 1.1 Review of Prior Work 1.2 Addressing the Research Gap 2 Research Approach 2.1 Social Science Theories 2.2 ABM Implementation 3 Results 3.1 Experiment 1: Effect of Contagion and Fear on Violence 3.2 Experiment 2: Sensitivity Analysis—System-Level Factors of Restraint () 4 Conclusions and Future Work References Global News Sentiment Analysis 1 Introduction 2 Related Work 3 Methodology and Experimentation 4 Results 4.1 Initial Observation 4.2 Comparison of Four Sentiment Analysis Tools 4.3 Three Article Processing Approaches 5 Discussion 6 Conclusion References An Agent-Based Model of Social Fabric Seen as an Emergent Behavior 1 Introduction 2 Social Fabric 3 Modeling Social Fabric in Cities 3.1 Conceptual Model 3.2 Agent-Based Model 4 Case Study: The Miramar Region 4.1 Mobility Patterns and Its Relationship with Perception 4.2 Encounters 5 Conclusion References Deep Agent: Studying the Dynamics of Information Spread and Evolution in Social Networks 1 Introduction 2 Challenge Problem Description 3 Methodology 3.1 Deep Agent Framework: Architecture and Analysis 3.2 Agent-Based Models 4 Dataset Description 4.1 Evaluation Events and Metrics 5 Experimental Results 6 Conclusion and Future Work References Electoral College: Emergent Battlegrounds An Agent-Based Model of Campaign Behavior Change with District Allocation of Electors 1 Introduction 2 Background 3 Methodology 4 The Model 5 Validation 6 Results 7 Discussion and Conclusions References Social Viscosity, Fluidity, and Turbulence in Collective Perceptions of Color: An Agent-Based Model of Color Scale Convergence 1 Introduction: Social Viscosity, Fluids, and Turbulence 2 Social Meaning and the Analysis of Proximity 2.1 Meaning and Viscosity in Color Agents 3 An Agent-Based Model of Color Proximity 4 Case Study: Nucleation in Circular Color Scales 4.1 Methods 4.2 All-to-all Interactions 4.3 N Nearest-Neighbor Interactions 5 Discussion 6 Conclusions and Further Work References Inside the Mind of the Nonfiler: An Agent-Based Modeling Approach 1 Introduction 1.1 Brief Overview of Nonfilers and IRS Enforcement/Outreach 1.2 Current IRS Nonfiler Research 1.3 Past Uses of Agent-Based Models (ABMs) at RAAS 2 Methods 2.1 Agent-Based Model Overview 2.2 Econometric Modeling 3 Data 4 Results 4.1 Agent-Based Model 5 Discussion 5.1 Agent-Based Model 6 Conclusions and Future Work 6.1 Future Work Appendix—Overview, Design Concepts and Details, and Decision Making (ODD + D) Protocol Documentation of the Nonfiler Decision Model Purpose Entities, State Variables, Scales Process Overview and Scheduling Design Concepts Initialization Input Data Submodels References Capturing the Effects of Gentrification on Property Values: An Agent-Based Modeling Approach 1 Introduction 2 Related Work 3 Model Design 3.1 Data 3.2 Environment 3.3 Agent Classes 3.4 Gentrification Calculation 3.5 Verification 4 Results 4.1 Gentrification by Supply 4.2 Gentrification by Demand 5 Discussion and Further Works 5.1 Main Results 6 Conclusion References Negative Influence Gradients Lead to Lowered Information Processing Capacity on Social Networks 1 Introduction 2 Related Work 3 Method 4 Results 5 Discussion 6 Conclusion References A Complex-Systems Agenda for Influencing Policy Studies 1 Introduction 1.1 Purpose 1.2 Our Definitions 1.3 Additional Discussion of Definitions 2 Worldview 2.1 System Thinking 2.2 Confronting Complexity and Wicked Problems 2.3 Decision-Making as a Continuing and Messy Process 2.4 The Values and Objectives that Drive Decisions 2.5 Knowledge, Uncertainty, and Disagreement 2.6 Moderation in the Search for Control 3 Basis for Reasoning and Inference 3.1 Theory, Data, Association, and Causation 3.2 Types of Theory 4 Analytic Style 4.1 Analysis and Reductionism, Both Good and Bad 4.2 Character of Analysis: Quantitative Versus Qualitative, and Matters of Rigor 4.3 Confronting Uncertainty and Disagreement 4.4 Comparing Options 4.5 Changing the Questions Asked of Simulation 5 Character of Models and Model-Based Analysis 5.1 Different Classes of Model 5.2 Purpose of Models and Related Issues of Validity 5.3 Measures of Outcome 5.4 Predict and Act Versus Anticipate Possibilities and Act Adaptively 6 Conclusions 6.1 Implications for Teaching Policy Studies 6.2 Implications for Computational Social Science References Wealth Dynamics in the Presence of Network Structure and Primitive Cooperation 1 Introduction 2 Model 2.1 Mathematical Formulation 2.2 Implementation 3 Verification 3.1 Agent and Proto Life History 3.2 Gini Coefficient History 4 Analysis 4.1 Gini Coefficient 4.2 Life Expectancy 4.3 Interpretations, Conjectures, and Next Steps 5 Conclusion and Future Work References Social Primitives: Exploring Spark of Life Collective Behavior in Agent-Based Models 1 Introduction 2 Examples of Agent-Based Social Primitives 2.1 Ring World 2.2 Conway—Game of Life 2.3 Complexity Adaptive Systems—Page and Miller 2.4 Hammond—Tobacco and Policy—Low-Dimensional and High-Dimensional Models 3 Taxonomy of Social Primitives 3.1 Agent Description 3.2 Coding Agents and the Challenge of Implicit Assumptions 3.3 Environment Description 3.4 Coding the Environment and the Challenge of Implicit Assumptions 4 Coding a Social Primitives Model: The SPECscape Model 4.1 Agent to Agent Interaction 4.2 Agent-Environment Interaction 4.3 Anthropomorphic Crossover 5 Running the SPECscape Model 5.1 SPECscape Model and Gini Index 5.2 Four Landscape Starting Points 5.3 The Effect of Exogenous Shocks at Various Stages of Model Evolution 5.4 Models with and without Proto-institutions 6 Conclusion 6.1 Elements of Collective Function 6.2 Determining Anthropomorphic Leaps 6.3 Determining a Parameter-Based Measure of Model Complexity 6.4 Possibilities for Increasing the Complexity of SPECscape Model Appendix References Capturing the Production of Innovative Ideas: An Online Social Network Experiment and ``Idea Geography'' Visualization 1 Introduction 2 Online Social Network Experiment 2.1 Experimental Procedure 2.2 Collective Design Tasks 3 Data Analysis Methods 3.1 Doc2Vec 3.2 Principal Component Analysis 3.3 Idea Geography 4 Results 4.1 Experiment Session I: Catch Phrase Design 4.2 Experiment Session II: Story Design 5 Discussions 6 Conclusions References An Agent-Based Model to Understand a Simple Theory of Mind: Belief Representation Systematic Approach (BRSA) 1 Introduction 1.1 Theory of Mind 1.2 Minimal Theory of Mind 1.3 Belief Representation Systematic Approach 2 Mental State Model 2.1 Mental States (MS) 2.2 Egocentric Agents (Control Agents) 2.3 Minimal Theory of Mind Agents (MinToM Agents) 2.4 Simple Theory of Mind Agents (Infer Agents) 3 Results 3.1 The Cost and Resources of ToM 4 Discussion 4.1 Simple ToM and the BRSA 4.2 Minimal Theory of Mind and the BRSA 4.3 ToM Ability and Agents’ Performances 4.4 Presumptions of Mental States (Biases) 5 Conclusions Appendix Overview Purpose State Variables and Scales Process Overview and Scheduling Design Concepts Details Initialization Input Submodels References Decision-Adjusted Modeling for Imbalanced Classification: Predicting Rooftop Solar Panel Adoption in Rural Virginia 1 Introduction 2 Related Work 3 Methodology 3.1 Data 3.2 Decision-Adjusted Model 4 Case Study and Results 4.1 Model Comparison 4.2 True Postitive Rate (TPR) + True Negative Rate (TNR) 4.3 Comparison Among Different Decision-Adjusted Models 4.4 Model Coefficients 5 Discussions References