Analyzing User and Group Behavior during Elections: Coordination, Bots and Misinformation. [Virtual Talk]

Fri, 29 Sep 2023 11:00:00 +0000

Abstract:

[ENG]

The unprecedented growth in social media use and large-scale information collection pose new threats, but also offer new opportunities. Modeling and managing complex interactive systems requires the analysis of social and technological signals to gain new insights into human society and individual behavior. Online social networks play an essential role in our access to information and are a good proxy for studying population-level behavior patterns and individual-level predictions. In this talk, I will present my research analyzing various account behaviors, from social bots spreading misinformation to coordinated activities during political campaigns and elections. I will also present the dataset we collected and the system we developed for the #Secim2023 project, which we used to study the recent Turkish presidential elections. I will also briefly discuss the other projects we are doing in the VRL lab.

[ESP]

El crecimiento sin precedentes en el uso de redes sociales y la recolección de información a gran escala plantean nuevas amenazas, pero también ofrecen nuevas oportunidades. Modelar y gestionar sistemas interactivos complejos requiere el análisis de señales sociales y tecnológicas para obtener nuevas perspectivas sobre la sociedad humana y el comportamiento individual. Las redes sociales en línea desempeñan un papel esencial en nuestro acceso a la información y son un buen proxy para estudiar patrones de comportamiento a nivel de población y predicciones a nivel individual. En esta charla, presentaré mi investigación analizando varios comportamientos de cuentas, desde bots sociales que difunden desinformación hasta actividades coordinadas durante campañas políticas y elecciones. También presentaré el conjunto de datos que recopilamos y el sistema que desarrollamos para el proyecto #Secim2023, que utilizamos para estudiar las recientes elecciones presidenciales turcas. También discutiré brevemente los otros proyectos que estamos realizando en el laboratorio VRL. (Traducido por ChatGPT).

Speaker Bio

Dr. Onur Varol is an Assistant Professor at the Sabanci University Faculty of Engineering and Natural Sciences and Principal Investigator at the VIRAL Lab. His research focuses on developing techniques to analyze online behaviors to improve individual well-being and address societal problems using online data. He is awarded by The Turkish Science Academy to Young Scientist Awards (2022) and Research Incentive Award by METU Parlar Foundation (2023). He also received TUBITAK 2247-D National Leader Researcher Grant in 2022. Prior to joining Sabanci University, he was a postdoctoral researcher at Northeastern University at the Center for Complex Network Research. He completed his PhD in Informatics at Indiana University, Bloomington (USA). His thesis focuses on the analysis of manipulation and threats on social media and he was awarded the 2018 University Distinguished Ph.D. Dissertation Award. He has developed a system called Botometer to detect social bots on Twitter and his team ranked top 3 worldwide at the 2015 DARPA Bot Detection Challenge. His research published in prestigious venues such as International Conference of Web and Social Media (ICWSM), Nature Communications, Nature Human Behavior,World Wide Web (WWW) conference, and Communications of the ACM.

Beyond Core-Periphery: Uncovering the Impacts of Scientific Networks on Resources and Recognition. [Virtual Talk]

Fri, 08 Sep 2023 11:00:00 +0000

Abstract:

[ENG]

We discuss two projects that quantify the consequences of our network location for our access to resources and rewards in science. First, we investigate the dynamics of international science through the lens of national citation preferences, introducing a novel measure to characterize these processes and enabling a network analysis of global knowledge production. Our results challenge the conventional core-periphery narrative of global science, instead uncovering several fragmented communities of knowledge production that are rapidly evolving and constraining the diffusion of ideas across international borders. Second, we quantitatively model the complex dynamics of philanthropic grant making in the US. Using publicly available tax records, we build the scientific funding network (925,335 grants between 69,000 institutions, totaling over $208 billion) to reveal that funders support geographically close recipients, and that grant-giving relationships become increasingly entrenched over time. These principles combine with high levels of bipartite clustering to empower a predictive model of future funding relationships. Together, these projects shed light on the barriers and opportunities for the equitable distribution of scientific resources and recognition, ultimately guiding policy recommendations to foster more inclusive and impactful scientific endeavors.

[ESP]

Discutimos dos proyectos que cuantifican las consecuencias de nuestra ubicación en la red en relación con nuestro acceso a recursos y recompensas en la ciencia. En primer lugar, investigamos la dinámica de la ciencia internacional a través del prisma de las preferencias de citación nacionales, introduciendo una medida novedosa para caracterizar estos procesos y permitiendo un análisis de red de la producción de conocimiento global. Nuestros resultados desafían la narrativa convencional del núcleo y la periferia en la ciencia global, descubriendo en su lugar varias comunidades fragmentadas de producción de conocimiento que evolucionan rápidamente y limitan la difusión de ideas a través de las fronteras internacionales. En segundo lugar, modelamos cuantitativamente la dinámica de la concesión de subvenciones filantrópicas en EE.UU. Utilizando registros fiscales públicos, construimos la red de financiamiento científico (925,335 subvenciones entre 69,000 instituciones, con un total de más de $208 mil millones) para revelar que los financiadores apoyan a los beneficiarios geográficamente cercanos y que las relaciones de concesión de subvenciones se vuelven cada vez más arraigadas con el tiempo. Estos principios, junto con altos niveles de agrupación bipartita, dan lugar a un modelo predictivo de futuras relaciones de financiamiento. Juntos, estos proyectos arrojan luz sobre las barreras y oportunidades para la distribución equitativa de recursos y reconocimientos científicos, guiando finalmente recomendaciones de políticas para fomentar esfuerzos científicos más inclusivos e impactantes. (Traducido por ChatGPT).

Speaker Bio

Alex Gates is a network scientist in the School of Data Science at the University of Virginia where he directs the Connected Data Hub. His research focuses on the Science of Science to analyze and model how organizational structure and strategic decisions impact innovation, creativity, and success. His work has been featured in top journals including Nature, PNAS, and The Journal of Machine Learning. https://alexandergates.net/

Science of Science | CRiSS-LAB Chile

Analyzing User and Group Behavior during Elections: Coordination, Bots and Misinformation. [Virtual Talk]

Abstract:

Speaker Bio

Beyond Core-Periphery: Uncovering the Impacts of Scientific Networks on Resources and Recognition. [Virtual Talk]

Abstract:

Speaker Bio