| Author: | Wang, Meng |
| Title: | Music analysis by time series data mining |
| Advisors: | Chan, Keith C. C. (COMP) |
| Degree: | M.Sc. |
| Year: | 2015 |
| Subject: | Data mining. Musical analysis -- Data processing. Hong Kong Polytechnic University -- Dissertations |
| Department: | Department of Computing |
| Pages: | ix, 78 leaves : illustrations ; 30 cm |
| Language: | English |
| Abstract: | With booming of Internet, the exploration of information has already been a hot topic in many fields. Traditional Database technology and data query methods are not able to deal with this large amount of data effectively and efficiently. So, Data Mining technology rises in response to the proper time and develops rapidly. Data means information in this era, and mining the data in higher and deeper layers can accelerate the transactions of the information, which has far-reaching impacts in nearly every fields in daily life. In the reality, most of the data is related with time, getting data in time order and analyzing, processing time series data has more meanings in the current situation. Moreover, in the recent years, an interesting analysis of music in time series data mining has emerged. This is a new area, which has not have many researches in it. The transformation of the pitch into numbers, the scientific study of the regularity and the analysis for the general structure have attracted more and more attentions. This dissertation focuses on pitch analysis from the Bach chorales, which has not have many concerns yet. The pitches from Bach chorales are analyzed without the consideration of duration, fermata and other characteristics. They can be treated as time series data, which are observations collected at the equal time intervals. Precisely, the pitch series is clustered with k-means clustering method to mining the general structure and distribution of the pitches in the soprano lines of 100 Bach chorales. The aim of the clustering is to generate the labels for data, which are useful for the further analysis and prediction. Moreover, this dissertation implements deeper music analysis on the pitches with the time series ARIMA model and exponential smoothing (ES) model, which has been conducted on the solid foundation of the clustering process. And, the comparisons between these two models will be completed, which ARIMA model is better for the accuracy of the whole data set, whereas ES model is good at fitting the peak values. The time series analysis and data mining of musical data can help audiences understand the structure and emotion behind the note better. In some certain degrees, the scientific perspectives of music analysis can promote the generation of melody. |
| Rights: | All rights reserved |
| Access: | restricted access |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| b27800234.pdf | For All Users (off-campus access for PolyU Staff & Students only) | 3.18 MB | Adobe PDF | View/Open |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/7844