Machine Learning Of Melodies Through Hierarchical Self-Organizing Maps

Hean, Edwin Law Hui (2010) Machine Learning Of Melodies Through Hierarchical Self-Organizing Maps. Masters thesis, University of Multimedia.

Full text not available from this repository.
Official URL: http://vlib.mmu.edu.my/diglib/login/dlusr/login.ph...

Abstract

In this thesis, discussions are made about several questions in music informatics through the design and development of a system to generate melodies and the system utilizes the Hierarchical Self-Organizing Map (HSOM), an unsupervised neural network model that is commonly used in tasks of data visualization and clustering, as a memory store of melodies. More specifically, the thesis tackles the problem of Melody Generation which is still a wide open research area in music informatics. Various methods that make use of techniques like machine learning, encoded rules, evolutionary computation, statistical analysis and so on have been proposed over the years and this thesis borrows ideas from various research to formulate a system inspired by the Memory Prediction Framework and built using the HSOM.

Item Type: Thesis (Masters)
Subjects: Q Science > Q Science (General)
Divisions: Faculty of Computing and Informatics (FCI)
Depositing User: Ms Suzilawati Abu Samah
Date Deposited: 12 Apr 2012 03:30
Last Modified: 12 Apr 2012 03:30
URI: http://shdl.mmu.edu.my/id/eprint/3491

Actions (login required)

View Item View Item