以Web标签为基础的相似歌曲研究

J4 ›› 2012, Vol. 47 ›› Issue (5): 53-58.

以Web标签为基础的相似歌曲研究

刘璇¹,许洁萍^1*,陈捷²

1. 中国人民大学信息学院, 北京 100872; 2. 北京师范大学信息科学与技术学院, 北京100875

收稿日期:2011-11-30 出版日期:2012-05-20 发布日期:2012-06-01
通讯作者: 许洁萍(1966- ),女,副教授,研究方向为多媒体信号处理,多媒体内容分析与检索,汉语语音标准研究.Email:xjieping@ruc.edu.cn
作者简介:刘璇(1985- ),女,硕士研究生,研究方向为多媒体信号处理. Email:rucliuxuan@hotmail.com
基金资助:
国家自然科学基金资助项目(60873098)

Music similarity research based on the Web tag

LIU Xuan¹, XU Jie-ping^1*, CHEN Jie²

1. Department of Computer Science and Technology, Renmin University of China, Beijing 100872, China;
2.College of Information Science and Technology, Beijing Normal University, Beijing 100875, China

Received:2011-11-30 Online:2012-05-20 Published:2012-06-01

摘要/Abstract

摘要：

以Last.fm音乐网站上的用户标签为特征进行歌曲相似性计算研究,提出了一种基于Web标签挖掘的歌曲相似计算方法。从音乐网站Last.fm上抽取了用户标签作为音乐语义特征,通过潜在语义分析(latent semantic analysis, LSA)方法对歌曲语义向量进行降维,并利用改进的K-means算法,根据音乐歌曲间的相似度对歌曲分类,完成了歌曲相似性计算。实验结果表明本文提出的基于LSA音乐语义标签模型的相似计算能取得较好的效果。

关键词: 标签;歌曲相似;LSA;降维;Web

Abstract:

A method was proposed for music classification using web tags on web mining, and the user tags made from Last.fm were used as features to study the music similarity. The web tags extracted from Last.fm were the music semantic feature. The Latent Semantic Analysis (LSA) was used for dimension reduction, and finally, according to the similarity between music, clustering results were obtained by the improved K-means. The experimental results show that the proposed method in this paper can get a better result in music classification.

Key words: tag; music similarity; LSA; dimension reduction; Web

刘璇1,许洁萍1*,陈捷2. 以Web标签为基础的相似歌曲研究[J]. J4, 2012, 47(5): 53-58.

LIU Xuan1, XU Jie-ping1*, CHEN Jie2. Music similarity research based on the Web tag[J]. J4, 2012, 47(5): 53-58.

参考文献

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed