Network inequality describes the formation and evolution of inequalities and biases that are caused by the connectivity patterns between the agents. Those inequalities and biases range from inequality of visibility, perception biases, algorithmic unfairness, and persistence of inequalities over time.

Understanding the generation of structural biases is not only a matter of theoretical interest. Many studies have addressed the effect of prestige and hierarchy in hiring, urban economies, or science. Moreover, recent works have found that algorithmic processes and recommendation systems based on relational structures can reinforce a bias through their learning process. In turn, adequate algorithmic designs could help decrease inequalities.

These data-driven and theoretical studies are inherently multidisciplinary, connecting fields such as computer science, physics, and social science. This satellite will be an excellent opportunity to get a holistic picture of structural inequalities by bringing together experts from various areas, including theoretical network science, anthropology, analytical social science, experimental psychology, fairness community in computer science, complex systems, and computational social science.

14:05 - 14:35: Emőke-Ágnes Horváth, Gender Inequality in Science: The Role of Networks

Despite recent efforts, the gender gap in science persists, for instance, in terms of citations, research funding and career success. In this talk, I will discuss our work on gender bias in scholars’ online visibility across various research areas and levels of success. Studying millions of scholars, we uncover the network-related factors associated with visibility and find no characteristics that consistently predict women's success online. I will also demonstrate how the bias in visibility impacts future citations, especially for intersectional coauthor teams. Finally, to enable the development of effective interventions, I will discuss differences in women's and men's self-promotion on social media. (Based on joint work with Orsolya Vásárhelyi, Hao Peng, Julia Barnett, Igor Zakhlebin, Misha Teplitskiy, Daniel Romero and Staša Milojević.)

14:40 - 14:55: Akrati Saxena, Diverse Link Prediction in Complext Networks

In real-world complex networks, understanding the dynamics of their evolution has been of great interest to the scientific community. Predicting non-existent but probable links is an essential task of social network analysis (SNA) as the addition or removal of the links over time leads to network evolution. In recent years, designing fairness-aware methods has received much attention in various domains, including machine learning, natural language processing, and information retrieval. However, understanding structural bias and inequalities in social networks and designing fairness-aware methods for various research problems in social network analysis have not received much attention. In this talk, I will first discuss link prediction methods and highlight how the structural bias of social networks impacts the fairness of link prediction. Next, I will discuss approaches to encounter structural biases for fair and diverse link prediction. I will cover one method in-depth for diverse link prediction using NodeSim network embedding that efficiently captures the diverse neighborhood while keeping the more similar nodes closer in the context of a given node. This will provide insights on how fairness can be incorporated in SNA. The talk will be concluded with a research vision to bridge the gap between fairness and SNA.

15:00 - 15:15: Takahiro Yabe, Behavioral changes during the pandemic decreased income diversity of urban encounters

Diversity of physical encounters in urban environments is known to spur economic productivity while also fostering social capital. However, mobility restrictions during the pandemic have forced people to reduce urban encounters, raising questions about the social implications of behavioral changes. Here, we study how individual income diversity of urban encounters changed during the pandemic, using a large-scale, privacy-enhanced mobility dataset of more than one million anonymized mobile phone users in US cities across three years spanning before and during the pandemic. We find that the diversity of urban encounters has substantially decreased during the pandemic and has persisted through late 2021, even though aggregated mobility metrics have recovered to pre-pandemic levels. Counterfactual analyses show that behavioral changes including lower willingness to explore new places further decreased the diversity of encounters in the long term. Our findings provide implications for managing the trade-off between the stringency of COVID-19 policies and the diversity of urban encounters as we move beyond the pandemic.

15:20 - 15:50: Talal Rahwan, Global Inequality in the Academy and the World Wide Web

The talk consists of two parts, the first of which focuses on inequality in academia. In terms of gender, we find that women are underrepresented on editorial boards: only one in seven editors, and one in twelve editors-in-chief, is a woman. Moreover, over the past decades, women have been consistently underrepresented among editors in every single discipline apart from Sociology. In terms of racial and geographical inequality, we find that most countries in Asia, Africa, and South America have fewer editors than would be expected based on their share of authorship. Focusing on U.S.-based scientists reveals that Blacks are the most underrepresented. As for the time spent between the submission and acceptance of a manuscript, we find that Black authors spend the longest time under review. Finally, analyzing citation rates of U.S.-based papers reveals that Black and Hispanic scientists receive fewer citations compared to White scientists doing similar research.

The second part of the talk focuses on digital inequality. The World Wide Web empowers people in developing regions by eradicating illiteracy, supporting women, and generating economic opportunities. However, their reliance on limited bandwidth and low-end phones leaves them with a poor browsing experience. We sent participants to 56 countries to measure global variation in web-browsing experience, revealing significant inequality in mobile data cost and page load time. We also find that popular webpages are increasingly tailored to high-end phones, thereby exacerbating this inequality. Our solution, Lite-Web, makes webpages faster to load and easier to process on low-end phones. Evaluating Lite-Web on the ground reveals that it transforms the browsing experience of Pakistani villagers with low-end phones to that of Dubai residents with high-end phones. These findings call attention from researchers and policy makers to mitigate digital inequality.

15:55 - 16:10: Sam Zhang, Minority group size moderates inequity-reducing strategies in homophilic networks

Members of minority groups often occupy less central positions in networks, which translates into a lack of access to essential information, social capital, and financial resources. This marginalization in relation to the majority group can arise from simple structural features of social networks: the tendency of individuals to connect to own in-group (homophily) and the preference for connecting to popular individuals in the network (preferential attachment). Yet in real life, groups of actors exist at the boundaries of minority and majority group memberships whose presence affects the visibility of the minority group. By modeling a social system with these boundary groups, we illuminate the hidden role of the minority group size in moderating (even non-linearly) the benefit of these intermediating groups. Our results suggest that interventions on structural inequities on networks can depend sensitively on the relative sizes of the groups involved. We observe the increasing difficulty of a minority group to achieve parity as the group shrinks: both external strategies (increasing allies) and internal strategies (increasing incorporation) require a higher relative threshold of participation before equity is reached.

16:15 - 16:30: Fabian Baumann, From polarization to ideologies: modeling opinion formation on dynamic networks

The prevalence of echo chambers and opinion polarization in various sociopolitical contexts and discussions on social media has raised concerns regarding their potential impact on the dissemination of misinformation and the openness of debates. In light of these concerns, we propose a model that incorporates the dynamics of radicalization as a reinforcing mechanism resulting from strongly biased information exchange between individuals. Drawing inspiration from empirical research on social interaction dynamics, our model considers agents with different level of activities and a tendency to interact with like-minded individuals (homophily).

By comparing our model's outcomes with empirical data on polarized debates observed on Twitter and the American National Election Survey (ANES), we find that it qualitatively reproduces the relationship between users' engagement and opinions, as well as the segregation of opinions within the interaction network. Furthermore, when considering multiple topics simultaneously, our approach offers insights into the formation of ideological states, where opinions on different topics exhibit strong correlations.