IEICE TRANSACTIONS on Information
Pairwise Similarity Normalization Based on a Hubness Score for Improving Cover Song Retrieval Accuracy
Summary :
A hubness-score based normalization of the pairwise similarity is proposed for the sequence-alignment based cover song retrieval. The hubness, which is the tendency of some data points in high-dimensional data sets to link more frequently to other points than the rest of the points from the set, is widely-known to deteriorate the information retrieval accuracy. This paper tries to relieve the performance degradation due to the hubness by normalizing the pairwise similarity with a hubness score. Experiments on two cover song datasets confirm that the proposed similarity normalization improves the cover song retrieval accuracy.
- Publication
- IEICE TRANSACTIONS on Information Vol.E105-D No.5 pp.1130-1134
- Publication Date
- 2022/05/01
- Publicized
- 2022/02/21
- Online ISSN
- 1745-1361
- DOI
- 10.1587/transinf.2021EDL8075
- Type of Manuscript
- LETTER
- Category
- Music Information Processing
Authors
Jin S. SEO
Gangneung-Wonju National University
Keyword
Latest Issue
Copyrights notice of machine-translated contents
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Cite this
Copy
Jin S. SEO, "Pairwise Similarity Normalization Based on a Hubness Score for Improving Cover Song Retrieval Accuracy" in IEICE TRANSACTIONS on Information,
vol. E105-D, no. 5, pp. 1130-1134, May 2022, doi: 10.1587/transinf.2021EDL8075.
Abstract: A hubness-score based normalization of the pairwise similarity is proposed for the sequence-alignment based cover song retrieval. The hubness, which is the tendency of some data points in high-dimensional data sets to link more frequently to other points than the rest of the points from the set, is widely-known to deteriorate the information retrieval accuracy. This paper tries to relieve the performance degradation due to the hubness by normalizing the pairwise similarity with a hubness score. Experiments on two cover song datasets confirm that the proposed similarity normalization improves the cover song retrieval accuracy.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2021EDL8075/_p
Copy
@ARTICLE{e105-d_5_1130,
author={Jin S. SEO, },
journal={IEICE TRANSACTIONS on Information},
title={Pairwise Similarity Normalization Based on a Hubness Score for Improving Cover Song Retrieval Accuracy},
year={2022},
volume={E105-D},
number={5},
pages={1130-1134},
abstract={A hubness-score based normalization of the pairwise similarity is proposed for the sequence-alignment based cover song retrieval. The hubness, which is the tendency of some data points in high-dimensional data sets to link more frequently to other points than the rest of the points from the set, is widely-known to deteriorate the information retrieval accuracy. This paper tries to relieve the performance degradation due to the hubness by normalizing the pairwise similarity with a hubness score. Experiments on two cover song datasets confirm that the proposed similarity normalization improves the cover song retrieval accuracy.},
keywords={},
doi={10.1587/transinf.2021EDL8075},
ISSN={1745-1361},
month={May},}
Copy
TY - JOUR
TI - Pairwise Similarity Normalization Based on a Hubness Score for Improving Cover Song Retrieval Accuracy
T2 - IEICE TRANSACTIONS on Information
SP - 1130
EP - 1134
AU - Jin S. SEO
PY - 2022
DO - 10.1587/transinf.2021EDL8075
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E105-D
IS - 5
JA - IEICE TRANSACTIONS on Information
Y1 - May 2022
AB - A hubness-score based normalization of the pairwise similarity is proposed for the sequence-alignment based cover song retrieval. The hubness, which is the tendency of some data points in high-dimensional data sets to link more frequently to other points than the rest of the points from the set, is widely-known to deteriorate the information retrieval accuracy. This paper tries to relieve the performance degradation due to the hubness by normalizing the pairwise similarity with a hubness score. Experiments on two cover song datasets confirm that the proposed similarity normalization improves the cover song retrieval accuracy.
ER -
English
