Linear Regression-Based Model Adaptation for Personalized Music Emotion Recognition
Date Issued
2014
Date
2014
Author(s)
Chen, Yu-An
Abstract
Personalization techniques can be applied to address the subjectivity issue of music emotion recognition, which is important for music information retrieval. However,
achieving satisfactory accuracy in personalized music emotion recognition for a user is difficult because it requires an impractically huge amount of annotations from the user. In this thesis, a linear regression based method is proposed to personalize a music emotion model in an online learning fashion. In addition, two parameter tying strategies are employed to improve the efficiency of personalization by modifying closely related model parameters together. An empirical justification of the design of parameter tying strategies are given, and comprehensive experiments conducted on several datasets showed the effectiveness of the proposed method.
Subjects
音樂
情緒辨識
個人化
機器學習
線性迴歸
模型參數類聚
Type
thesis
File(s)![Thumbnail Image]()
Loading...
Name
ntu-103-R01942128-1.pdf
Size
23.32 KB
Format
Adobe PDF
Checksum
(MD5):51cff32f404fcdf34f9b278393d5e780