CEU Department of Network and Data Science Research Contributes to Pandemic Response on Multiple Fronts

With the COVID-19 pandemic reaching the duration of a year, researchers at Central European University (CEU) have made significant knowledge contributions to inform pandemic response. A density of such projects has emerged from CEU’s Department of Network and Data Science (DNDS), which is led by Professor Janos Kertesz.

Professor Janos Kertesz, Photo: CEU

“Whenever disease, information or trends are spreading, they spread on networks. So when the pandemic started, we felt that this was integral to our research,” explains Kertesz. DNDS Associate Professor Marton Karsai and Visiting Professor Julia Koltai have also been deeply immersed in COVID-related research, including collaborations across CEU departments as well as with governmental agencies and Hungarian Universities working on pandemic response. “These are all interdisciplinary problems and studying them needs to be a cooperative undertaking, from within and outside the department,” notes Kertesz.

Contributing to research efforts, Karsai and Koltai developed the Hungarian Data Provider Questionnaire (MASZK), which serves as a major source of data for several projects. The survey is a voluntary anonymous online questionnaire to dynamically record the age stratified interaction patterns of people together with other information about their habits, employment, mindset and opinion on topics like vaccination, trust and misinformation. The MASZK data collection and epidemiological modeling operation is coordinated by the University of Szeged and has been providing the Hungarian government with analysis of the possible courses that the pandemic might take.

This survey, with more than 400,000 responses, has reached over 2.3% of the Hungarian population. Additionally, a representative monthly data collection campaign collects responses from 1000-1500 people via a phone survey. This analyzed data informs epidemic modeling and is used to advise the COVID Action Team in Hungary. First results of their research have been recently shared in a pre-print paper.

Associate Professor Marton Karsai, Photo: CEU

Karsai, in addition to being part of the MASZK team with Koltai, directs epidemic modelling research at the Rényi Institute of Mathematics, Hungarian Academy of Sciences as part of the DYNASNET ERC Synergy project together with László Lovász. Their research team analyzes geometric meta-population networks and applies models of Hungarian data to understand the impact of various early pandemic spreading scenarios on the final outcome of the epidemic.

“My team is focusing on how to estimate the changes of local social interaction patterns via longitudinal surveys, while also studying how social networks are reorganized or mobility patterns are changed during emergency situations like lockdowns,” notes Karsai. “In terms of modeling, we are interested in how the spread of an epidemic depends on the spatially embedded structure of a mobility network and on the location of its initial infection seeds.” Karsai emphasizes how network science provides central tools to understand the dynamics and the outcome of epidemic processes as network epidemics constitutes a field studying how the structure and dynamics of social interactions or mobility patterns influence epidemic transmission locally or on a global level.

Additionally, in collaboration with the UNICEF Office of Innovation and the DSTI Office of the Sierra Leone Government, Karsai and his PhD student Ludovico Napoli are conducting research to understand how the social networks of people reorganize during COVID lockdowns. This research is performed on a large scale anonymized national mobile phone communication dataset aiming to understand how people of different socioeconomic classes can adapt to pandemic restrictions, which in turn may change segregation patterns in the social fabric.

Looking to other pandemic-influenced social phenomena, Karsai with his PhD student Rafiazka Hilman in collaboration with the UNICEF Office of Innovation are studying how COVID-19 lockdowns change the mixing of different socioeconomic classes in urban settings in the US. The research is based on a large anonymized individual level mobility dataset and nationwide spatially detailed census maps, with special focus on neighborhoods where local outbreaks of COVID were stronger.

Visiting Professor Julia Koltai, Photo: Daniel Vegel

Furthermore, Koltai in collaboration with Associate Professor of Gender Studies and Co-Director of the CEU Democracy Institute Éva Fodor, turned to the pandemic’s uneven impact on gender, investigating how lockdown influenced the division of household labor in Hungary. Their paper, titled “The impact of COVID-19 on the gender division of childcare work in Hungary,” published in European Societies, examines the impact of school and childcare facility closures. During the pandemic, an increased volume of childcare was supplied by individual households without significant institutional help. In analyzing the gendered divisions in childcare duties, they found that women’s contributions grew significantly more than men’s and the gap between men and women has increased in absolute work hour terms. The data suggests that in Hungary, the pandemic increased gender inequality the most among the highest educated.

Koltai and Fodor’s latest pandemic-related project focuses on gendered attitudes toward willingness to get the vaccination on the representative data of the above-mentioned Hungarian Data Provider Questionnaire. “Earlier research suggests, that in Hungary women are significantly less likely to vaccinate themselves than men,” comments Koltai. In their latest research, they are reconstructing the complex interactions of social characteristics and processes which lead to an anti-vaccination attitude, thus pointing to those dimensions which are different in the case of men and women. Their results can be useful in the communication of vaccine propagation.

On the topic of information flows, Kertész, together with his PhD student Hao Cui investigated the effect of the pandemic on the Chinese microblog Sina Weibo. For their forthcoming paper, “Attention dynamics on the Chinese social media Sina Weibo during the COVID-19 pandemic“ they focused on the dynamics of attention and identified major factors necessary to reach masses of people on online social networks, which can be crucial to saving lives. “Understanding the dynamics of public attention is an extremely important topic nowadays as the question is no longer how to get information but how to select information. During a pandemic, it is important that people are informed and that is possible only if their attention is captured,” explains Kertesz.

Finally, Kertesz, with another PhD student, András Borsos, in collaboration with the Complexity Science Hub Vienna and the Hungarian National Bank, are investigating how a major economic shock, like the one caused by the pandemic, propagates through the network of economy. Through their model and analysis, they locate key players on the company level and investigate anticipated effects of various interventions. Through this research, they are developing a tool to provide policy-making advice for selecting the most appropriate measure in promoting recovery. 

From pandemic modeling and social network shifts, to information flows and changing divisions of household labor, DNDS’s timely contributions of dimensional scholarship extend beyond the sphere of academia to meaningfully contribute to society’s understanding and response to the pandemic.