Social, Cultural, and Behavioral Modeling: 13th International Conference, SBP-BRiMS 2020, Washington, DC, USA, October 18–21, 2020, Proceedings (Lecture Notes in Computer Science Book 12268)

This book constitutes the proceedings of the 13th International Conference on Social, Cultural, and Behavioral Modeling, SBP-BRiMS 2020, which was planned to take place in Washington, DC, USA. Due to the COVID-19 pandemic the conference was held online during October 18–21, 2020. The 33 full papers presented in this volume were carefully reviewed and selected from 66 submissions. A wide number of disciplines are represented including computer science, psychology, sociology, communication science, public health, bioinformatics, political science, and organizational science. Numerous types of computational methods are used, such as machine learning, language technology, social network analysis and visualization, agent-based simulation, and statistics. Preface Organization Contents Beyond Words: Comparing Structure, Emoji Use, and Consistency Across Social Media Posts 1 Introduction 2 Background 2.1 Content Analysis of Social Media 2.2 Analysis of Emoji Use 3 Measuring Document Structure, Emoji Use, and Consistency 3.1 Document Structure, Content Structure, and Emoji Spans 3.2 Attributes of Emojis in a Document 3.3 Emoji Position, Order, and Repetition 3.4 Measure of Consistency 3.5 Clustering by Structure, Content, and Emoji Use 3.6 Clustering Users Based on Consistency 4 Experiment Results and Discussion 4.1 Experimental Setup 4.2 Distributions of Consistency 4.3 Analysis of Structures for Document, Content, and Emoji Use 4.4 Clustering Users Based on Consistency 4.5 Behavior Traits and Composition of Users in Clusters 5 Conclusions and Future Work References Bot Impacts on Public Sentiment and Community Structures: Comparative Analysis of Three Elections in the Asia-Pacific 1 Introduction 2 Methods 2.1 Data Collection 2.2 Bot Detection 2.3 Bot Activities 2.4 Network Influence 2.5 Sentiment Analysis 2.6 Echo-Chamberness 2.7 Causal Inference 3 Results 3.1 Bot Prevalence 3.2 Bot Influence 3.3 Causal Inference on Bot Impacts 4 Discussion References Understanding Colonial Legacy and Environmental Issues in Senegal Through Language Use 1 Introduction 2 Background and Related Work 2.1 Natural Language Processing and Word Embeddings 2.2 Biases and Stereotypes Captured by Word Embeddings 2.3 Senegal 3 Training Word Embeddings 3.1 Text Processing Steps 3.2 Validation of the Word Embeddings 4 Social Issue Analysis with Word Embeddings 4.1 Legacy of French Assimilation 4.2 Environmental Problems 4.3 Past - Present - Future in Senegal 5 Discussion 6 Conclusion References Deploying System Dynamics Models for Disease Surveillance in the Philippines 1 Introduction 2 Methods 2.1 Data Collection 2.2 System Dynamics Model of Dengue Fever 2.3 Simulated Annealing for Parameter Estimation 2.4 Virtual Experiments for Evaluating Interventions 3 Results 3.1 Model Initialization 3.2 Simulated Interventions 4 Discussion References MDR Cluster-Debias: A Nonlinear Word Embedding Debiasing Pipeline 1 Introduction 2 Related Work 2.1 Bias Definition 2.2 Debiasing Word Embeddings 3 Our Debiasing Pipeline 3.1 Post-processing Procedure 3.2 Cluster-Debias 4 Evaluation Methods 5 Experiments 5.1 Results 6 Conclusion References Modeling Interventions for Insider Threat 1 Introduction 2 Background 3 Model Details 4 Virtual Experiments 4.1 Experimental Design 4.2 Virtual Experiment 1 4.3 Virtual Experiment 2 4.4 Virtual Experiment 3 5 Discussion 6 Conclusion References Validating Social Media Monitoring: Statistical Pitfalls and Opportunities from Public Opinion 1 Introduction 2 Threats to Statistical Conclusion Validity on Social Media 2.1 Threat 1: Heterogeneity of Units 2.2 Threat 2: Extraneous Variance in Experimental Setting 2.3 Threat 3: Inaccurate Estimation of Effect Sizes 2.4 Threat 4: Incorporate Statistical Controls from Public Opinion 3 Conclusion Validity on Social Media: A Checklist References Lying About Lying on Social Media: A Case Study of the 2019 Canadian Elections 1 Introduction 2 Related Work 3 Methods 3.1 Data 3.2 Bot Identification 3.3 Detecting Targets of #FakeNews Accusations 4 Results and Analysis 4.1 Network Structure 4.2 #FakeNews Analysis 4.3 #NotABot Analysis 5 Discussion 6 Conclusion and Future Work References Breadth Verses Depth: The Impact of Tree Structure on Cultural Influence 1 Introduction 1.1 Related Work 2 Model 3 Methods 4 Results 4.1 Polarised, Homogenised and Mixed Links in a Model of Social Influence 4.2 Polarisation at Different Tree Levels 5 Conclusion References Optimizing Attention-Aware Opinion Seeding Strategies 1 Introduction 2 Model 3 Methods 4 Results 5 Discussion 6 Conclusions and Future Work References Polarizing Tweets on Climate Change 1 Introduction 2 Data Collection and Method 2.1 Data Collection 2.2 Method 3 Results 3.1 Topics of Discussion 3.2 User Accounts and Conversations 4 Discussion References Characterizing Sociolinguistic Variation in the Competing Vaccination Communities 1 Introduction 2 Dataset 2.1 Data Collection 2.2 Community Detection 2.3 Timeline Extraction 2.4 Data Statistics 3 Methodology 3.1 Linguistic Analysis 3.2 Network Analysis 3.3 Evaluation 4 Results and Discussion 4.1 Linguistic Analysis 4.2 Network Analysis 5 Limitations and Future Work 6 Conclusion References On Countering Disinformation with Caution: Effective Inoculation Strategies and Others that Backfire into Community Hyper-Polarization 1 Introduction 2 Model Description 2.1 Model Parameters 2.2 Rules of Interaction 3 Experiments and Results 3.1 Collective-Thought 3.2 Opinion Profile 3.3 Experiment 1: Effect of and on 3.4 Experiment 2: Effect of on 4 Conclusions and Future Work References Homicidal Event Forecasting and Interpretable Analysis Using Hierarchical Attention Model 1 Introduction 2 Literature Review 3 Problem Statement 4 Model Architecture 4.1 Word-to-Sentence Embedding 4.2 Sentence Encoder Module 4.3 Article Encoder Module 5 Experimental Evaluation 5.1 Dataset Description 5.2 Feature Engineering and Data Pre-processing 5.3 Comparative Models 6 Results and Discussion 7 Conclusion References Development of a Hybrid Machine Learning Agent Based Model for Optimization and Interpretability 1 Introduction 2 Learning-Driven Actor-Interpreter State Representation Model (LAISR) 3 Tactical Air and Ground Warfare Experiment 3.1 Game Simulation Layer 3.2 Actor/Interpreter Layer 4 Results 5 Discussion and Future Work References Canadian Federal Election and Hashtags that Do Not Belong 1 Introduction 2 Background and Motivation 3 Canadian Election Twitter Data 4 Methodology 5 Results and Discussion 6 Conclusion References Group Formation Theory at Multiple Scales 1 Introduction 2 Testing Existing Theory at Multiple Scales 2.1 Dataset Description 2.2 Group Cohesiveness 2.3 Sociometer Theory 2.4 Terror Management Theory and Mortality Salience 3 Conclusions References Towards Agent Validation of a Military Cyber Team Performance Simulation 1 Introduction 2 Background 3 Model Overview 3.1 Agent Behaviors 3.2 Agent Output Data 4 Simulation Setup and Results 5 Model Output Validation Design Considerations 5.1 Cyber Force Agent Validation Design Considerations 5.2 Terrain Agent Validation Design Considerations 6 Conclusion References Developing Graph Theoretic Techniques to Identify Amplification and Coordination Activities of Influential Sets of Users 1 Introduction 2 Literature Review 3 Proposed Methodology 3.1 Local Analysis-User Level 3.2 Sets of Users Analysis- Network Level 3.3 Influential Sets of Users Validation Level 4 Experimental Results 5 Conclusion and Discussion Appendix I:Zachary Karate Club Network [6] References Detecting Online Hate Speech: Approaches Using Weak Supervision and Network Embedding Models 1 Introduction 2 Related Work 3 Methodology 3.1 Weak Supervision Model 3.2 Multilayer Network Embedding 4 Experiments and Results 4.1 Dataset Description 4.2 Weak Supervision Model 4.3 User Classification 5 Conclusion and Discussion References Critical Spatial Clusters for Vaccine Preventable Diseases 1 Introduction 2 Preliminaries 2.1 Disease Spread on a Social Contact Network 2.2 Criticality 2.3 Problem Formulation 3 Our Approach 4 Experimental Results 5 Related Work 6 Conclusions References Multi-cause Discrimination Analysis Using Potential Outcomes 1 Introduction 2 Preliminaries 2.1 Potential Outcome Framework 2.2 Structural Causal Model and Intervention 3 Modeling Multi-cause Discrimination 3.1 Problem Formulation 3.2 The Deconfounder Algorithm 3.3 Inverse Probability of Treatment Weighting 4 Empirical Evaluation 4.1 Synthetic Data 4.2 Adult Dataset 5 Conclusion References Twitter Is the Megaphone of Cross-platform Messaging on the White Helmets 1 Introduction 2 Data Collection and Processing 3 Mentions of Digital Platforms in Twitter Messages 4 Media Co-sharing Patterns 4.1 Tweet/Cascade-Sharing Network 4.2 User-Sharing Network 5 Summary References Learning Behavioral Representations from Wearable Sensors 1 Introduction 2 Related Work 3 Methods 3.1 Measuring Distance 3.2 Learning Representations 4 Data 5 Results 5.1 Clustering 5.2 Prediction 6 Conclusion References The Rise and Fall of Humanitarian Citizen Initiatives: A Simulation-Based Approach 1 Introduction 2 Methodology 3 Model Specification 4 Scenario Development and Evaluation 4.1 Humanitarian Crisis Onset Scenario 4.2 Humanitarian Crisis Stabilization Phase Scenario 4.3 Scenario Development as Evaluation 5 Discussion and Future Research References Developing an Epidemiological Model to Study Spread of Toxicity on YouTube 1 Introduction 2 Literature Review 2.1 Identifying Toxicity on Social Media 2.2 Epidemiological Models 3 Methodology 3.1 Data Collection 3.2 Spread of Toxicity 3.3 Model Validation 4 Conclusions and Future Work References Predicting Student Flight Performance with Multimodal Features 1 Introduction 2 Related Work 3 Experiments 3.1 Procedure 3.2 Equipment 3.3 Data Collection 4 Multimodal Features 4.1 Gaze Data 4.2 Flight Control Data 4.3 Knowledge Mastery 5 Machine Learning Models 6 Results and Discussion 7 Conclusion and Future Work References A Game-Transformation-Based Framework to Understand Initial Conditions and Outcomes in the Context of Cyber-Enabled Influence Operations (CIOs) 1 Introduction 2 Background 3 Methodology 4 The Game Transformation-Based Framework 4.1 Level One 4.2 Level 2 5 Discussion and Conclusion References The Human Resource Management Parameter Experimentation Tool 1 Introduction 2 Background 3 Methodology 4 Results 5 Summary and Further Work References Utilizing Python for Agent-Based Modeling: The Mesa Framework 1 Introduction 1.1 Mesa’s Design Goals 2 Architecture and Usage 2.1 The Model 2.2 Model Analysis 2.3 Model Visualization 2.4 Creating a Model 3 Applications and Extensions of Mesa 4 Conclusion References Strategic Information Operation in YouTube: The Case of the White Helmets 1 Introduction 2 Dataset 3 Spoofing 4 Astroturfing? 5 Conclusions References Artifacts of Crisis: Textual Analysis of Euromaidan 1 Introduction 2 Prior Work 3 Methodology and Data 3.1 Legislation 3.2 Blogs 3.3 Twitter 4 Results 4.1 Legislation 4.2 Blogs 4.3 Twitter 4.4 Data Stream Timing 5 Conclusions and Future Work References Modeling Decisions from Experience Among Frequent and Infrequent Switchers via Strategy-Based and Instance-Based Models 1 Introduction 2 Problem Dataset 3 The Models 3.1 The Round-Wise and Summary Models 3.2 The Instance-Based Model 4 Method 4.1 Dependent Variables 4.2 Model Calibration 5 Human Results 6 Model Results 7 Discussion and Conclusions References Author Index This book constitutes the proceedings of the 13th International Conference on Social, Cultural, and Behavioral Modeling, SBP-BRiMS 2020, which was planned to take place in Washington, DC, USA. Due to the COVID-19 pandemic the conference was held online during October 18-21, 2020. The goal of this conference is to build this new community of social cyber scholars by bringing together and fostering interaction between members of the scientific, corporate, government, and military communities interested in understanding, forecasting, and impacting human sociocultural behavior. It is the charge of this community to build this new science, its theories, methods, and its scientific culture in a way that does not give priority to either social science or computer science, and to embrace change as the cornerstone of the community. As the papers in this volume show, people, theories, methods, and data from a wide number of disciplines are represented including computer science, psychology, sociology, communication science, public health, bioinformatics, political science, and organizational science. Numerous types of computational methods are used, such as machine learning, language technology, social network analysis and visualization, agent-based simulation, and statistics