BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//UNIFR/WEBMASTER//NONSGML v1.0//EN
CALSCALE:GREGORIAN
BEGIN:VEVENT
DTSTART;VALUE=DATE:20260610T140000
DTEND;VALUE=DATE:20260610T140000
UID:19847@agenda.unifr.ch
DESCRIPTION:The ranking of entities in complex competitive systems based on pairwise interactions is a\nstandard problem in network science. This work uses the men’s professional tennis circuit\nas a case study, modelling pairwise match outcomes as a directed network, to evaluate\ndifferent ranking methods. This assessment is challenging, as players’ intrinsic ability is not\ndirectly observable or quantifiable in reality. Historical ATP data were analysed to extract\ncalibration parameters governing player lifecycles, aging, and the distribution of talent.\nBased on these elements, a dynamic simulation model was developed to generate synthetic\ntennis circuits in which the latent skill score of each player is known. Comparing the various\nranking metrics against this ground truth reveals varied performances. Centrality-based\nmethods, such as PageRank, struggle to identify the top-tier players of the circuit. The\nprobabilistic maximum likelihood approach based on the Bradley-Terry model rivals the\nofficial ATP points system, although the latter remains, overall, the most robust method for\nreconstructing the underlying player hierarchy.
SUMMARY:Ranking Nodes in Competition-Driven Directed Networks
CATEGORIES:Autre
LOCATION:PER 08\, 2.73\, Chemin du Musée 3\, 1700 Fribourg
URL;VALUE=URI:https://agenda.unifr.ch/e/fr/19847
END:VEVENT
END:VCALENDAR