Modularity-Based Fairness in Network Communities

Manolis, Konstantinos

Modularity-Based Fairness in Network Communities

dc.contributor.author	Manolis, Konstantinos	en
dc.date.accessioned	2024-03-22T09:03:11Z
dc.date.available	2024-03-22T09:03:11Z
dc.identifier.uri	https://olympias.lib.uoi.gr/jspui/handle/123456789/37053
dc.identifier.uri	http://dx.doi.org/10.26268/heal.uoi.16764
dc.rights	CC0 1.0 Universal	*
dc.rights.uri	http://creativecommons.org/publicdomain/zero/1.0/	*
dc.subject	Social Networs, Community Detection, Community Fairness	en
dc.title	Modularity-Based Fairness in Network Communities	en
dc.type	masterThesis	en
heal.abstract	In this thesis, we study the fairness of community structures in networks from a group-based perspective. Specifically, we assume that individuals in a social network belong to different groups based on the value of one of their sensitive attributes, such as their age, gender, or race. We view community fairness as the lack of discrimination towards any of the groups. For simplicity, let us assume that nodes belong to two groups, the blue and the red group. We introduce three fairness metrics. The first metric, termed balance-fairness, equitably represents communities by ensuring an equal distribution of red and blue nodes in each community. The second, termed modularity-fairness, refines the notion of modularity to demand equal intracommunity connectivity for the groups. The third metric, termed diversity-fairness, promotes intra-community edges between nodes of different color thus addressing the filter-bubble phenomenon. We have modified the Louvain algorithm, a well-known community detection algorithm, to produce communities that are both well-connected and fair. We present an extensive evaluation using several real-world and synthetic networks. The goal of our evaluation is twofold: (1) to study the fairness of communities in networks and the causes of unfairness and (2) to evaluate the effectiveness of our fairness-enhanced Louvain algorithm.	en
heal.academicPublisher	Πανεπιστήμιο Ιωαννίνων. Πολυτεχνική Σχολή. Τμήμα Μηχανικών Ηλεκτρονικών Υπολογιστών και Πληροφορικής	el
heal.academicPublisherID	uoi	el
heal.access	free	el
heal.advisorName	Pitoura, Evaggelia	en
heal.classification	Social Networks
heal.committeeMemberName	Lykas, Aristidis	en
heal.committeeMemberName	Tsaparas, Panagiotis	en
heal.dateAvailable	2024-03-22T09:04:12Z
heal.fullTextAvailability	true
heal.identifier.secondary	Social Networks	el
heal.language	en	el
heal.publicationDate	2024-02-23
heal.recordProvider	Πανεπιστήμιο Ιωαννίνων. Πολυτεχνική Σχολή	el
heal.secondaryTitle	Modularity-Based Fairness in Network Communities	en
heal.type	masterThesis	el
heal.type.el	Μεταπτυχιακή εργασία	el
heal.type.en	Master thesis	en