Developing a user-specific affect model for classifying music based on timbral content

Date of Publication

2010

Document Type

Bachelor's Thesis

Degree Name

Bachelor of Science in Computer Science

Subject Categories

Computer Sciences

College

College of Computer Studies

Department/Unit

Computer Science

Thesis Adviser

Merlin Suarez

Defense Panel Member

Jocelynn Cu
Rhia Trogo

Abstract/Summary

The relationship between timbre and emotion has not yet been explored in-depth. Specifically, a user-specific affect model for classifying music by timbral content has yet to be created. Such a model would provide an insight into the specific of the relationship between timbre and emotion. Several steps are taken on the construction of such a model, namely the corpora building, the classification, and the prototyping. Preliminary data mining and data analysis was done with the help of Weka. The type and number of the instruments used was also taken into account, and normalization and principal components analysis was also used in analyzing the data. The verification and classification of the emotional response from all the subjects and analyzing the timbral content of those pieces. Some pieces were found whose timbral content induces consistent emotional responses from all the subjects.

Abstract Format

html

Language

English

Format

Print

Accession Number

TU18465

Shelf Location

Archives, The Learning Commons, 12F, Henry Sy Sr. Hall

Physical Description

1v. various foliations ; 28 cm. + 1 computer optical disc ; 4 3/4 in.

Keywords

Music--Emotion; Tone color (Music)

This document is currently not available here.

Share

COinS