Abstract
We proposed a method to classify songs in the Million Song Dataset according to song genre. Since songs have several data types, we trained sub-classifiers by different types of data. These sub-classifiers are combined using both classifier authority and classification confidence for a particular instance. In the experiments, the combined classifier surpasses all of these sub-classifiers and the SVM classifier using concatenated vectors from all data types. Finally, the genre labels for the Million Song Dataset are provided.
| Original language | English (US) |
|---|---|
| Title of host publication | SIGIR'12 - Proceedings of the International ACM SIGIR Conference on Research and Development in Information Retrieval |
| Pages | 1083-1084 |
| Number of pages | 2 |
| DOIs | |
| State | Published - 2012 |
| Event | 35th Annual ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2012 - Portland, OR, United States Duration: Aug 12 2012 → Aug 16 2012 |
Other
| Other | 35th Annual ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2012 |
|---|---|
| Country | United States |
| City | Portland, OR |
| Period | 8/12/12 → 8/16/12 |
Fingerprint
Keywords
- classifier combination
- song genre classification
ASJC Scopus subject areas
- Information Systems
Cite this
Genre classification for million song dataset using confidence-based classifiers combination. / Hu, Yajie; Ogihara, Mitsunori.
SIGIR'12 - Proceedings of the International ACM SIGIR Conference on Research and Development in Information Retrieval. 2012. p. 1083-1084.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Genre classification for million song dataset using confidence-based classifiers combination
AU - Hu, Yajie
AU - Ogihara, Mitsunori
PY - 2012
Y1 - 2012
N2 - We proposed a method to classify songs in the Million Song Dataset according to song genre. Since songs have several data types, we trained sub-classifiers by different types of data. These sub-classifiers are combined using both classifier authority and classification confidence for a particular instance. In the experiments, the combined classifier surpasses all of these sub-classifiers and the SVM classifier using concatenated vectors from all data types. Finally, the genre labels for the Million Song Dataset are provided.
AB - We proposed a method to classify songs in the Million Song Dataset according to song genre. Since songs have several data types, we trained sub-classifiers by different types of data. These sub-classifiers are combined using both classifier authority and classification confidence for a particular instance. In the experiments, the combined classifier surpasses all of these sub-classifiers and the SVM classifier using concatenated vectors from all data types. Finally, the genre labels for the Million Song Dataset are provided.
KW - classifier combination
KW - song genre classification
UR - http://www.scopus.com/inward/record.url?scp=84866596498&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84866596498&partnerID=8YFLogxK
U2 - 10.1145/2348283.2348480
DO - 10.1145/2348283.2348480
M3 - Conference contribution
AN - SCOPUS:84866596498
SN - 9781450316583
SP - 1083
EP - 1084
BT - SIGIR'12 - Proceedings of the International ACM SIGIR Conference on Research and Development in Information Retrieval
ER -