Title

Automatic music genre classification system

Date of Publication

2008

Document Type

Bachelor's Thesis

Degree Name

Bachelor of Science in Computer Science

College

College of Computer Studies

Department/Unit

Computer Science

Thesis Adviser

Nelson Marcos

Defense Panel Member

Joel Ilao
Rafael Cabredo

Abstract/Summary

To address the problems of manual classification, the proponents created a system that will automatically classify music into their genres. The genres considered are blues, classical, hip hop, jazz, pop and rock. The system compares different feature extraction to come up with good algorithm contributions. The system makes represent make use feature extraction algorithms to compute for the feature vectors, which represent the data to be classified. The features used are timbre, rhythm and pitch. K-Nearest Neighbor which is the classification algorithm used, utilize these feature vectors to compute for decision boundaries that separate each genre and classify the music to its corresponding genre. The system achieved 69.42% using timbre, 41.25% using rhythm, 32.75% using pitch and 72.83% using the combination of the three features and K = 6. Blues and classical songs were classified more accurately than other genres.

Abstract Format

html

Language

English

Format

Print

Accession Number

TU14647

Shelf Location

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

This document is currently not available here.

Share

COinS