POSTER SESSION NETSCI 2023

The Women in Network Science society is organising a poster session at the NetSci 2023 conference to be held in Vienna, Austria, featuring work by women and non-binary researchers in network science. 

The vibrant collection of posters will be showcased on Tuesday July 11th, 2023  throughout the day, with discussions happening over the coffee breaks (10h30-11h00 and 15h30-16h00). 

Everyone is welcome to join! 

Abstracts

Raphtory: the temporal networks engine for Python and Rust

Naomi Arnold*, Network Science Institute, Northeastern University London

Academic Mobility as a Driver of Productivity: A Gender-centric Approach

Mariana Macedo*, CCL-ANITI, University of Toulouse

Ana Maria Jaramillo, Network Inequality Group, CSH Vienna

Ronaldo Menezes, BioComplexLab, University of Exeter

pSTEM fields (Physical Sciences, Technology, Engineering and Mathematics) are known for showing a gender imbalance favouring men. This imbalance can be seen at several levels, including in university and industry, where men are the majority of the posts. Academic success is partly dependent on the value of the researchers’ co-authorship networks. One of the ways to enrich one’s network is through academic movement; the change of institutions in search of better opportunities within the same country or internationally. In this paper, we look at the data for one specific pSTEM field, Computer Science, and describe the productivity and co-authorship patterns that emerge as a function of academic mobility. We find that women and men both benefit from national and international mobility, women who never change affiliations over their career are rarely well-cited or highly productive, and women are not well-represented in the overall top-ranking researchers.

Configuration models for directed hypergraphs

Yanna Kraakman*, University of Twente

Network null models are broadly used in network analysis. We propose a null model for directed hypergraphs, which samples uniformly for certain classes of directed hypergraphs. The model is based on the edge-swapping model for directed graphs.

Social Interactions in Public Elementary Schools’ Neurodiverse Classrooms

Patricia Soto-Icaza

Melanie Oyarzún*, Universidad del Desarrollo Chile 

Tamara Yaikin

Mirla Arcos

Cristian Candia

Carlos Rodríguez-Sickert

Pablo Billeke 

Knowing how to live with others who are different from us is crucial for the development of satisfactory social relationships, particularly during childhood. From an inclusive perspective, studying social interactions is even more relevant when it comes to schools’ neurodiverse classrooms, where children with special educational needs engage and participate together with other students in formative activities.  Evidence has described children with Autism Spectrum Disorder (ASD), and Attention Deficit and Hyperactivity Disorders (ADHD) displayed difficulties in peer functioning. To clarify this issue, we studied the role of ASD and ADHD, in social relationships in educational institutions with a National School Integration Program for neurodiverse students in Chile. We designed a computational game in which children must select 10 named classmates to play and allocate 15 stars among them. Thus, for each classroom, a social network was built based on the number of individual selections and received stars. A total of 625 children between 6 and 11 years old participated in the study, from 6 elementary schools.  The results showed that children with special educational needs were selected less and received fewer stars than typically developed children and the difference is even larger considering children with ASD. These results suggest that certain neurodevelopmental conditions are related to the formation of social interaction networks in elementary classrooms and their impact on school coexistence.

Co-Evolving Dynamics and Topology in a Coupled Oscillator Model of Resting Brain Function

Maria Pope*, Luddy School of Informatics, Indiana University

Dynamic models of ongoing BOLD fMRI brain dynamics and models of communication strategies have been two important approaches to understanding how brain network structure constrains function. However, dynamic models have yet to widely incorporate one of the most important insights from communication models: the brain may not use all of its connections in the same way or at the same time. Here we present a variation of a phase delayed Kuramoto coupled oscillator model that dynamically limits communication between nodes on each time step. An active subgraph of the empirically derived anatomical brain network is chosen in accordance with the local dynamic state on every time step, thus coupling dynamics and network structure in a novel way. We analyze this model with respect to its fit to empirical time-averaged functional connectivity, finding that, with the addition of only one parameter, it significantly outperforms standard Kuramoto models with phase delays. We also perform analyses on the novel time series of active edges it produces, demonstrating a slowly evolving topology moving through intermittent episodes of integration and segregation. We hope to demonstrate that the exploration of novel modeling mechanisms and the investigation of dynamics of networks in addition to dynamics on networks may advance our understanding of the relationship between brain structure and function.

Topological determinants of convergence rate in distributed averaging gossip algorithms

Christel Sirocchi*, Department of Pure and Applied Sciences, University of Urbino

Alessandro Bogliol, Department of Pure and Applied Sciences, University of Urbino

Gossip algorithms are message-passing schemes designed to compute averages and other global functions over networks through asynchronous and randomised pairwise interactions. Gossip-based protocols have garnered significant attention due to their ability to achieve robust, fault-tolerant communication while maintaining simplicity and scalability. However, the frequent propagation of redundant information makes them inefficient and resource intensive. Most previous works have been devoted to deriving performance bounds and developing faster algorithms tailored to specific structures. In contrast, this study focuses on characterising the effect of topological network features on performance so that faster convergence can be engineered by acting on the underlying network rather than the gossip algorithm. Numerical experiments identify the topological limiting factors and the most predictive structural features for each graph family and all graphs, providing guidelines for designing and maintaining resource-efficient networks. In addition, the study uncovers the predictive capabilities of local graph metrics, which can be computed in a distributed manner and at a low computational cost. In light of these findings, a novel distributed approach for predicting performance from the network topology is proposed. This approach empowers individual nodes to forecast the time or the number of interactions required to estimate the global average with the desired accuracy. Consequently, nodes can make informed decisions on their use of measured and estimated data while gaining awareness of the global structure of the network, as well as their role in it. Furthermore, this study showcases relevant applications of the proposed protocol, including outliers identification and performance evaluation in switching topologies.

Understanding the evolution of successful careers in professional tennis

Chiara Zappalà*, University of Catania

Success in tennis is a captivating yet complex phenomenon that has garnered minor research attention. In this study, we delve into the evolution of players' careers, aiming to unveil the predictors and mechanisms behind the emergence of top players. Despite existing studies on success in tennis, there is still limited understanding of the factors that contribute to it. Specifically, the role of early career stages and the impact of tournament attendance on players' trajectories remain intriguing factors requiring further investigation. To address these knowledge gaps, we employ a comprehensive approach combining network science and analysis of ATP tournament data. We introduce a novel measure of tournament prestige based on a co-attendance network of tournaments and eigenvector centrality. Through our analysis, we unveil insightful findings concerning the evolution of players' careers. Notably, we differentiate the impact of tournament attendance from players' performances, emphasizing the significance of the level of the tournament where players achieve their first win in characterizing top players. This research contributes to our understanding of success in tennis by shedding light on the role of early career stages and the interplay between tournament attendance and performance. Our findings offer valuable insights to researchers seeking to comprehend the dynamics of success in tennis.

Film industry networks: from gendered film production to festival programming

Vejune Zemaityte*, Media and Art School, Tallinn University 

The emergence of detailed datasets describing the dynamics of the international film industries has facilitated various computational analyses of the domain, including different applications of network methods (Lutter, 2015; Mourchid et al., 2019; Park et al., 2012). This poster showcases three network analyses of different film industry aspects, each situated in a unique context and performed under distinct methodological configuration. The first study offers a cross-country comparison of the gendered feature film production networks during 2006–2016. We use data about films and their key creatives (producers, directors, and writers) in three countries to draw unipartite directed producer–crew hiring networks that describe 344 films and 611 people in Australia, 304 films and 687 people in Sweden, and 1,446 films and 2,756 people in Germany. Each national film production industry features stark gender inequality, where a high proportion of men only work with other men. The second study concerns longitudinal gender dynamics in the short documentary clip (‘newsreel’) production industry of the Soviet Union. Based on data about 1,747 clips produced by 1,623 directors, cinematographers, text editors, and other crew during 1945–1992, we explore gendered labour relationships via unipartite directed director–crew networks. In contrast to the stark gender inequality found in the contemporary setting, the directorial work in the Soviet Union is split equitably between women and men, and women directors are very well-embedded within the production network structures as characterised by high degree centrality, especially during the 1960s. The last study uses network methods to unveil the structure of the international film festival circuit. From the bipartite film–festival network featuring 31,989 films programmed at 616 festivals, we project a unipartite undirected festival–festival network based on films shared between the programming of different events during 2009–2021. The network is highly interconnected contrary to a stratified and disunified festival circuit assumed in previous qualitative literature. The three introduced analyses showcase the applicability of network methods to the film industry domain in a multidisciplinary setting and invite further discussion.

*Speaker

Organizers

Gabriela Juncosa (Central European University, Austria)

Zsofia Zador (Northeastern University London, United Kingdom)

Alice Schwarze (Dartmouth College, USA)