Δυναμικές ανάπτυξης δικτύων ομάδων

Θεολογής, Νικόλαος; Theologis, Nikolaos

dc.contributor.author	Θεολογής, Νικόλαος	el
dc.contributor.author	Theologis, Nikolaos	en
dc.date.accessioned	2024-11-08T12:36:52Z
dc.date.available	2024-11-08T12:36:52Z
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/60403
dc.identifier.uri	http://dx.doi.org/10.26240/heal.ntua.28099
dc.description	Εθνικό Μετσόβιο Πολυτεχνείο--Μεταπτυχιακή Εργασία. Διεπιστημονικό-Διατμηματικό Πρόγραμμα Μεταπτυχιακών Σπουδών (Δ.Π.Μ.Σ.) "Επιστήμη Δεδομένων και Μηχανική Μάθηση"	el
dc.rights	Αναφορά Δημιουργού 3.0 Ελλάδα	*
dc.rights.uri	http://creativecommons.org/licenses/by/3.0/gr/	*
dc.subject	Affinity Bias	en
dc.subject	Informational Diversity	en
dc.subject	DeGroot Learning	en
dc.subject	Opinion Formation	en
dc.subject	Graph Theory	en
dc.title	Δυναμικές ανάπτυξης δικτύων ομάδων	el
dc.title	Network team growth dynamics	en
heal.type	masterThesis
heal.classification	Team Formation	en
heal.language	en
heal.access	free
heal.recordProvider	ntua	el
heal.publicationDate	2024-06-25
heal.abstract	Previous research has convincingly demonstrated that in organizational settings, teams characterized by a diverse range of information and perspectives tend to outperform their homogeneous counterparts. Despite this evidence, why do we frequently observe predominantly homogeneous teams in practice? One prevailing explanation posits that the advantages of informational diversity are in tension with affinity bias. To delve deeper into the implications of this conflict on team composition, we study a sequential model of team formation. In this model, individuals prioritize their team's performance, as measured by its ability to accurately predict future outcomes based on various features, while also considering the potential costs associated with interacting with teammates who employ different approaches to the prediction task. Our work extends this initial team formation model by adding an underlying graph structure that changes how both the accuracy of the team and the disagreement between team members are calculated. We study two different graph structures. The first is a random undirected graph for which we have the freedom of changing and adjusting the edges in order to reach the optimal cost, while the second is a rigid hierarchical pyramid structure in which the edges are fixed in place, allowing us only the freedom to optimally position the agents within the pyramid. These extensions keep the tension between informational diversity and affinity bias, which we aim to optimize either by ensuring the optimal connections within the team or by strategically positioning the team members.	en
heal.advisorName	Fotakis, Dimitris	en
heal.committeeMemberName	Pagourtzis, Aris	en
heal.committeeMemberName	Terzi, Evimaria	en
heal.academicPublisher	Εθνικό Μετσόβιο Πολυτεχνείο. Σχολή Ηλεκτρολόγων Μηχανικών και Μηχανικών Υπολογιστών	el
heal.academicPublisherID	ntua
heal.numberOfPages	74 σ.	el
heal.fullTextAvailability	false