Efficient Over-time Analysis of Social Network Data Sets

Analysis of over-time Social Network Data is extremely important in understanding how a given social discourse is evolving. This can include changes in network structure (the connections between agents) and in the emotions and words being communicated. In this session, we will focus on over-time analysis of changes in the network structure by making use of a variety of social network metrics. In part, the changes over time of key social network metrics can indicate that a significant change in the society is happening. We will explain how the ability to detect such changes can be enhanced through the use of advanced signal processing techniques. In addition, we will explore the use of social network metrics for extracting various types of social groups (often called the community structure). We will discus the computational problem that arises in the detection of groups, given dynamic network data. In particular, we will consider the challenge of detecting in time variant incremental data with as little delay as possible given vast quantities of network data.

Weaponized Information and the Science of Mental Immunity

The new science of mental immunity (I call it “cognitive immunology”) harbors a shift in perspective that could fundamentally rebalance our relationship with the vectors of cognitive contagion. In this talk, I’ll introduce you to the science, clarify the shift in perspective, and show why we need them both to combat weaponized (dis)information.

Detecting and analyzing hate speech with natural language processing

Social media and online forums have become the latest communication technologies to be used to target historically marginalized groups. Such hate speech limits participation in online communities, enables hateful ideologies to spread, and has been linked to real-world violence. In this talk, I will first discuss challenges in using natural language processing to develop automated methods for detecting hate speech, including building datasets and annotating this relatively rare, difficult-to-define phenomenon. I will then discuss possible solutions and future directions for computational hate speech analysis. These include moving beyond binary classification of hate speech toward identifying what speakers are doing with hate speech (spreading stereotypes, discussing violence, etc.). I will also discuss the potential for incorporating greater communicative and social network context in detecting and analyzing hate speech online.

Narratives and Networks of Online Hate

To understand and counter online hate, it is crucial to identify the narratives in which it is embedded and the networks in which it spreads. We characterize how online hate narratives specifically target various social identities, and measure the network features of online communities that propagate hate. Through this framework, we move beyond quantifying the prevalence of online hate, to a more holistic examination of its dynamics within and against social groups. Mapping narratives and networks further offers broader insight into the ways online hate is manipulated through information operations, suggesting concrete paths forward toward mitigating its impacts.

Temporal changes in coordination

Agent-Based Simulation of Reddit Discussions

Walk through of the process used to simulate Reddit communities and discussions using Construct as the framework for the simulations.

A network-based approach to learning dynamics between particles

Social media networks are dynamical systems whose state varies over time due to the cumulative interactions between entities. Like entities in social media networks, particles in real-world systems interact with one another because of underlying dynamics. The future state of such a system of particles can be predicted by learning the underlying dynamics that govern their interactions. This demonstration will walk you through the process of simulating, collecting, and learning the interactions/dynamics between particles in simple physical systems that can be modelled as networks.

What is "Pink Slime” and how can we detect it?

Illegitimate regional news sites controlled by national entities (“pink slime”) have been popping up to infiltrate local communities. This talk will discuss the rise of pink slime, its characteristics and impact, and how new sources can be detected using social network analysis.

Topic/Description

The effects of evolutionary adaptations on spreading processes in complex networks

In this talk, we will present recent advances in mathematical modeling and analysis of spreading processes over networks. Our focus will be on information and misinformation spread over social networks, but we will also talk about how these tools and concepts can be applied more broadly to other spreading processes including viral spread. Specifically, we will discuss the impact of mutations and evolutionary adaptations on the spreading process. Secondly, we will present results on spreading processes over multi-layer networks with an eye towards revealing the impact of multiple social media platforms on the speed and extent of information spread. 

Success in the Marketplace of IDeaS

Some ideas spread widely; others don't.  Whether an idea spreads has little relation to whether or not it is true.  In this workshop, I review the psychology literature on what makes ideas believable, memorable, and contagious, with the ultimate goal of creating a more informed society. 

Influencing truth: Exploring appeals that work when facts do not

We know that facts will likely not change the minds or behaviors of people who are committed to believing that vaccines can’t be trusted, that the election was stolen or that the earth is flat. But what best practices might we borrow from an industry whose very existence is based on influencing behavior? Advertising agencies and design firms have over the years employed techniques of awareness, influence and persuasion to encourage reticent audiences to do everything from prevent forest fires to rock the vote. Are there effective techniques used in advertising and design that can be applied to change the negative or neutral behaviors of reticent or untrusting audiences? We explore some of the best and most effective practices used in the most successful public service campaigns and evaluate their potential effect in the areas of vaccine hesitancy, election trust, and the spread of misinformation and disinformation.

Why and How to Research TikTok

This talk will be from a qualitative user studies perspective on social media research. Discussion will also include quantitative approaches and how they are useful in mixed-method designs.

Network Mechanisms of Polarization in Social Media

Two general approaches in network science have been applied to explain the growing polarization and extremism observed in social media platforms. One approach views polarization as diffusion processes in which a small portion of active users spread their skewed beliefs and attitudes to a larger number of their network neighbors online. Another approach focuses on the selection effect in social media where users who would have otherwise refrained from expressing contentious, extremist ideas offline add their emboldened voices to the polarized choir of like-minded social media users. In this session, we will first review research taking these approaches and then explore a third approach that focuses on how the transparency of interactions in social media can lead to relational constraints that limit users’ self-expression online, more so for the political moderates who are more likely connected to groups with opposing political views. The resulting relative paucity of moderate viewpoints leaves unchecked the extreme viewpoints of the more vocal minority, thereby skewing the observed opinion space towards increased polarization. After introducing this third approach, we will discuss strategies to measure users’ curtailed self-expression on Twitter, present recent empirical findings of this third approach, and discuss implications for informed democracy.

Applying Social Psychology to Cybersecurity

This talk will present the results of our work in understanding the role of social psychology in people's cybersecurity behaviors and in designing interventions that use social influence, such as the "Adulting 101" app and the "Safesea" app for Facebook. It will wrap up with SA-6 and SA-13 survey instruments that we've developed to measure security attitudes, plus some ongoing work.

Analysis of Chinese and Russian State Sponsored Information Operations

In this study, we analyze state-sponsored information operations in terms of the B.E.N.D framework and persuasion, identity, & morality. We analyze and compare multiple operations across time and between countries.