Creating & evaluating a music recommender system without access to multiple user data

Georgios, Pipilis; Γεώργιος, Πιπιλής

dc.contributor.author	Georgios, Pipilis	en
dc.contributor.author	Γεώργιος, Πιπιλής	el
dc.date.accessioned	2023-04-06T11:01:00Z
dc.date.available	2023-04-06T11:01:00Z
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/57509
dc.identifier.uri	http://dx.doi.org/10.26240/heal.ntua.25206
dc.description	Εθνικό Μετσόβιο Πολυτεχνείο--Μεταπτυχιακή Εργασία. Διεπιστημονικό-Διατμηματικό Πρόγραμμα Μεταπτυχιακών Σπουδών (Δ.Π.Μ.Σ.) "Επιστήμη Δεδομένων και Μηχανική Μάθηση"	el
dc.rights	Default License
dc.subject	Recommender system	en
dc.subject	Clustering	en
dc.subject	Spotify	en
dc.subject	Music recommendation	en
dc.subject	Content based filtering	en
dc.subject	Πρόταση μουσικής	el
dc.subject	Συστήματα προτάσεων	el
dc.subject	Ομαδοποίηση	el
dc.subject	Μηχανική Μάθηση	el
dc.subject	Σπότιφαι	el
dc.title	Creating & evaluating a music recommender system without access to multiple user data	en
heal.type	masterThesis
heal.secondaryTitle	Implementation and Testing on the Spotify Platform	en
heal.classification	Computer Science	en
heal.language	en
heal.access	free
heal.recordProvider	ntua	el
heal.publicationDate	2022-09-01
heal.abstract	A recommender system is a type of algorithm that provides personalized recommendations to users based on either their past behaviors or preferences or the properties of the content they consume. It is commonly used in e-commerce, streaming services, social media platforms, and other applications to enhance user experience and engagement. When it comes to music, recommender systems usually rely on vast amounts on user or track data in order to generate suggestions. This diploma thesis aims to explore the creation and evaluation of a full recommender system pipeline that does not rely on data from multiple users or bleeding edge computing resources in order to function. This is done through the exploration of a listener’s Spotify music history. The final algorithm, as well as the methods in which it is evaluated, will be compared to Spotify in order to evaluate how far one can reach without the need for extra resources.	en
heal.advisorName	Kollias, Stefanos	en
heal.committeeMemberName	Stamou, Georgios	en
heal.committeeMemberName	Kollias, Stefanos	en
heal.committeeMemberName	Voulodimos, Athanasios	en
heal.academicPublisher	Εθνικό Μετσόβιο Πολυτεχνείο. Σχολή Ηλεκτρολόγων Μηχανικών και Μηχανικών Υπολογιστών	el
heal.academicPublisherID	ntua
heal.numberOfPages	55 σ.	el
heal.fullTextAvailability	false