Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.668101
Title: Engaging with music retrieval
Author: Boland, Daniel
ISNI:       0000 0004 5365 3726
Awarding Body: University of Glasgow
Current Institution: University of Glasgow
Date of Award: 2015
Availability of Full Text:
Access through EThOS:
Access through Institution:
Abstract:
Music collections available to listeners have grown at a dramatic pace, now spanning tens of millions of tracks. Interacting with a music retrieval system can thus be overwhelming, with users offered ‘too-much-choice’. The level of engagement required for such retrieval interactions can be inappropriate, such as in mobile or multitasking contexts. Music recommender systems are widely employed to address this issue, however tend toward the opposite extreme of disempowering users and suffer from issues of subjectivity and confounds, such as the equalisation of tracks. This challenge and the styles of retrieval interaction involved are characterised in terms of user engagement in music retrieval, and the relationships between existing conceptualisations of user engagement is explored. Using listening histories and work from music psychology, a set of engagement-stratified profiles of listening behaviour are developed. A dataset comprising the playlists of thousands of users is used to contribute a user-centric approach to feature selection. The challenge of designing music retrieval for different levels of user engagement is first explored with a proof of concept, low engagement music retrieval system enabling users to casually retrieve music by tapping its rhythm as a query. The design methodology is then generalised with an engagement-dependent system, allowing users to denote their level of engagement and thus the specificity of their music queries. The engagement-dependent retrieval interaction is then explored as a component in a commercial music system. This thesis contributes the engagement-stratified profiles and metrics of listening behaviour, a corresponding design methodology for interaction, and presents a set of research and commercial applications for music retrieval.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.668101  DOI: Not available
Keywords: Z665 Library Science. Information Science ; ZA4050 Electronic information resources ; ZA4450 Databases
Share: