Main content

Evaluating Melodic Similarity Using Pairwise Sequence Alignments and Suffix Trees

Show simple item record

dc.contributor.advisor Calvert, David A. Wickland, David D. 2017-09-19T19:20:05Z 2017-09-19T19:20:05Z 2017-09 2017-09-13 2017-09-19
dc.description.abstract The music industry was the first mass-media industry to be transformed by the transition to digital. Today, music consumers have access to personal libraries, subscription-based libraries, and online content via web-media hosts or channels. Now more than ever, there are tremendous quantities of music available to the individual on a host of devices and platforms. With the explosion in access, there comes increasing demand to describe, index, retrieve, and interact with digital music. Consequently, there is increasing demand for novel and efficient ways to compare and manipulate digital music. The goal of investigating melodic similarity is to develop ways of comparing melodies based on structural elements or patterns present in the melody. This approach differs from other music similarity measures, which typically employ low-level features, or clustering based on associations. With robust and efficient ways of evaluating melodic similarities, we can develop new approaches to interacting with digital music. en_US
dc.language.iso en en_US
dc.rights Attribution-ShareAlike 2.5 Canada *
dc.rights.uri *
dc.subject symbolic similarity en_US
dc.subject melodic similarity en_US
dc.subject local alignment en_US
dc.subject global alignment en_US
dc.subject music similarity en_US
dc.subject suffix tree en_US
dc.subject cover song en_US
dc.subject longest common substring en_US
dc.subject pattern matching en_US
dc.subject sequence alignment en_US
dc.subject pairwise en_US
dc.subject exact matching en_US
dc.subject inexact matching en_US
dc.subject approximate matching en_US
dc.subject string matching en_US
dc.subject music information retrieval en_US
dc.subject MIR en_US
dc.title Evaluating Melodic Similarity Using Pairwise Sequence Alignments and Suffix Trees en_US
dc.type Thesis en_US Computer Science en_US Master of Science en_US School of Computer Science en_US
dc.rights.license All items in the Atrium are protected by copyright with all rights reserved unless otherwise indicated.

Files in this item

Files Size Format View Description
Wickland_David_201709_Msc.pdf 3.160Mb PDF View/Open MSc Thesis - revisions requested by Atrium made

This item appears in the following Collection(s)

Show simple item record

Attribution-ShareAlike 2.5 Canada Except where otherwise noted, this item's license is described as Attribution-ShareAlike 2.5 Canada