Web music emotion recognition based on higher effective gene expression programming
Web music emotion recognition based on higher effective gene expression programming
复制标题
基于高效基因表达编程的网络音乐情感识别
DOI：
10.1016/j.neucom.2012.06.041
复制
发表时间：
2013-04-01
期刊：
影响因子：
6
通讯作者：
Sun, Shouqian
中科院分区：
文献类型：
--
作者：
Zhang, Kejun;Sun, Shouqian
在本研究中，我们提出了一种更有效的算法，称为修正的基因表达式编程（RGEP），来构建音乐情感识别的模型。本文的主要工作如下：首先，描述了音乐情感识别的基本机制，并引入基因表达式编程（GEP）来处理音乐情感识别的模型构建问题。其次，本文提出了基于后向进化算法的RGEP，并分别用GEP、RGEP和SVM建立了音乐情感识别模型，实验结果表明，SVM、GEP和RGEP得到的模型都是令人满意的，与实验值吻合较好。最后，对这两种模型进行了比较，发现RGEP模型的分类精度优于GEP模型，处理时间比GEP模型减少近15%，比SVM模型减少一半，为解决音乐情感识别问题提供了一种新的有效途径;此外，由于处理时间对于大规模音乐信息检索问题至关重要，因此，RGEP可能会促进音乐信息检索技术的发展。(C)2012爱思唯尔有限公司版权所有。
In the study, we present a higher effective algorithm, called revised gene expression programming (RGEP), to construct the model for music emotion recognition. Our main contributions are as follows: firstly, we describe the basic mechanisms of music emotion recognition and introduce gene expression programming (GEP) to deal with the model construction for music emotion recognition. Secondly, we present RGEP based on backward-chaining evolutionary algorithm and use GEP, RGEP, and SVM to construct the models for music emotion recognition separately, the results show that the models obtained by SVM, GEP, and RGEP are satisfactory and well confirm the experimental values. Finally, we report the comparison of these models, and we find that the model obtained by RGEP outperforms classification accuracy of the model by GEP and takes almost 15% less processing time of GEP and even half processing time of SVM, which offers a new efficient way for solving music emotion recognition problems; moreover, because processing time is essential for the problem of large scale music information retrieval, therefore, RGEP might prompt the development of the music information retrieval technology. (C) 2012 Elsevier B.V. All rights reserved.