Responsible Link-Recommendations in Dynamic Environments
This proposal aims to develop, for the first time, new computational models to systematically evaluate 1) the long-term societal impacts of link-recommendation algorithms in online social networks and 2) design a new paradigm of l...
This proposal aims to develop, for the first time, new computational models to systematically evaluate 1) the long-term societal impacts of link-recommendation algorithms in online social networks and 2) design a new paradigm of link-recommenders that incentivize cooperation, collective action, and misinformation control.It is urgent to understand how algorithms used in online social media impact human behavioural dynamics given the widespread use of social media platforms and the evidence that they contribute to exacerbate radicalization, misinformation, and incite hate. This is a challenging endeavour. Online platforms are nowadays complex ecosystems where millions of humans influence each other while co-existing with AI algorithms. In this context, link-recommendation algorithms, used to recommend new connections to users, are ubiquitous. Such algorithms fundamentally affect how new connections are formed and the information users are exposed to. Governing online social networks requires understanding the impact of link-recommenders and how to adapt them to ensure long-term benefits.With RE-LINK, I aim to develop a new class of models to understand the impact of link-recommendations on social dynamics and, in turn, design a new paradigm of algorithms that balance short-term performance and long-term societal benefits. This will be achieved by developing agent-based models where the evolution of behaviours occurs over adaptive networks whose growth, in turn, follows the heuristics used by link-recommenders. I will resort to evolutionary game theory and stochastic population dynamics to formally study the stability of behaviours in this setting. I will use the modelling results to design new link-recommenders that contribute to stabilize positive social behaviours such as cooperation, collective action, and misinformation debunking. The developed algorithms will be evaluated with large-scale multi-agent simulations, online experiments, and real-world data.ver más
Seleccionando "Aceptar todas las cookies" acepta el uso de cookies para ayudarnos a brindarle una mejor experiencia de usuario y para analizar el uso del sitio web. Al hacer clic en "Ajustar tus preferencias" puede elegir qué cookies permitir. Solo las cookies esenciales son necesarias para el correcto funcionamiento de nuestro sitio web y no se pueden rechazar.
Cookie settings
Nuestro sitio web almacena cuatro tipos de cookies. En cualquier momento puede elegir qué cookies acepta y cuáles rechaza. Puede obtener más información sobre qué son las cookies y qué tipos de cookies almacenamos en nuestra Política de cookies.
Son necesarias por razones técnicas. Sin ellas, este sitio web podría no funcionar correctamente.
Son necesarias para una funcionalidad específica en el sitio web. Sin ellos, algunas características pueden estar deshabilitadas.
Nos permite analizar el uso del sitio web y mejorar la experiencia del visitante.
Nos permite personalizar su experiencia y enviarle contenido y ofertas relevantes, en este sitio web y en otros sitios web.