Improving music genre classification using collaborative tagging data

verfasst von: Ling Chen, Phillip Wright, Wolfgang Nejdl
Abstract: As a fundamental and critical component of music information retrieval (MIR) systems, music genre classification has attracted considerable research attention. Automatically classifying music by genre is, however, a challenging problem due to the fact that music is an evolving art. While most of the existing work categorizes music using features extracted from music audio signals, in this paper, we propose to exploit the semantic information embedded in tags supplied by users of social networking websites. Particularly, we consider the tag information by creating a graph of tracks so that tracks are neighbors if they are similar in terms of their associated tags. Two classification methods based on the track graph are developed. The first one employs a classification scheme which simultaneously considers the audio content and neighborhood of tracks. In contrast, the second one is a two-level classifier which initializes genre label for unknown tracks using their audio content, and then iteratively updates the genres considering the influence from their neighbors. A set of optimizing strategies are designed for the purpose of further enhancing the quality of the two-level classifier. Extensive experiments are conducted on real-world data collected from Last.fm. Promising experimental results demonstrate the benefit of using tags for accurate music genre classification.
Organisationseinheit(en): Forschungszentrum L3S
Externe Organisation(en): Georgia Institute of Technology
Typ: Aufsatz in Konferenzband
Seiten: 84-93
Anzahl der Seiten: 10
Publikationsdatum: 09.02.2009
Publikationsstatus: Veröffentlicht
Peer-reviewed: Ja
ASJC Scopus Sachgebiete: Computernetzwerke und -kommunikation, Software
Elektronische Version(en): https://doi.org/10.1145/1498759.1498812 (Zugang: Geschlossen)

BibTeX

@inproceedings{b65f23e673e34a6cb86c80238a3c3f8f,
title = "Improving music genre classification using collaborative tagging data",
abstract = "As a fundamental and critical component of music information retrieval (MIR) systems, music genre classification has attracted considerable research attention. Automatically classifying music by genre is, however, a challenging problem due to the fact that music is an evolving art. While most of the existing work categorizes music using features extracted from music audio signals, in this paper, we propose to exploit the semantic information embedded in tags supplied by users of social networking websites. Particularly, we consider the tag information by creating a graph of tracks so that tracks are neighbors if they are similar in terms of their associated tags. Two classification methods based on the track graph are developed. The first one employs a classification scheme which simultaneously considers the audio content and neighborhood of tracks. In contrast, the second one is a two-level classifier which initializes genre label for unknown tracks using their audio content, and then iteratively updates the genres considering the influence from their neighbors. A set of optimizing strategies are designed for the purpose of further enhancing the quality of the two-level classifier. Extensive experiments are conducted on real-world data collected from Last.fm. Promising experimental results demonstrate the benefit of using tags for accurate music genre classification.",
keywords = "Exploiting tag semantics, Music genre classification, Relaxation labelling",
author = "Ling Chen and Phillip Wright and Wolfgang Nejdl",
year = "2009",
month = feb,
day = "9",
doi = "10.1145/1498759.1498812",
language = "English",
isbn = "9781605583907",
series = "Proceedings of the 2nd ACM International Conference on Web Search and Data Mining, WSDM'09",
publisher = "Association for Computing Machinery (ACM)",
pages = "84--93",
booktitle = "Proceedings of the 2nd ACM International Conference on Web Search and Data Mining, WSDM'09",
address = "United States",
note = "2nd ACM International Conference on Web Search and Data Mining, WSDM'09 ; Conference date: 09-02-2009 Through 12-02-2009",
}