Please use this identifier to cite or link to this item:
https://open.uns.ac.rs/handle/123456789/5376
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Hara K. | en |
dc.contributor.author | Suzuki I. | en |
dc.contributor.author | Shimbo M. | en |
dc.contributor.author | Kobayashi K. | en |
dc.contributor.author | Fukumizu K. | en |
dc.contributor.author | Radovanović, Milan | en |
dc.date.accessioned | 2019-09-30T08:47:34Z | - |
dc.date.available | 2019-09-30T08:47:34Z | - |
dc.date.issued | 2015-06-01 | en |
dc.identifier.isbn | 9781577357025 | en |
dc.identifier.uri | https://open.uns.ac.rs/handle/123456789/5376 | - |
dc.description.abstract | Copyright © 2015, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved. Hubness has been recently identified as a problematic phenomenon occurring in high-dimensional space. In this paper, we address a different type of hubness that occurs when the number of samples is large. We investigate the difference between the hubness in highdimensional data and the one in large-sample data. One finding is that centering, which is known to reduce the former, does not work for the latter. We then propose a new hub-reduction method, called localized centering. It is an extension of centering, yet works effectively for both types of hubness. Using real-world datasets consisting of a large number of documents, we demonstrate that the proposed method improves the accuracy of knearest neighbor classification. | en |
dc.relation.ispartof | Proceedings of the National Conference on Artificial Intelligence | en |
dc.title | Localized centering: Reducing hubness in large-sample data | en |
dc.type | Conference Paper | en |
dc.identifier.scopus | 2-s2.0-84960130900 | en |
dc.identifier.url | https://api.elsevier.com/content/abstract/scopus_id/84960130900 | en |
dc.relation.lastpage | 2651 | en |
dc.relation.firstpage | 2645 | en |
dc.relation.volume | 4 | en |
item.grantfulltext | none | - |
item.fulltext | No Fulltext | - |
Appears in Collections: | Naučne i umetničke publikacije |
SCOPUSTM
Citations
14
checked on Feb 22, 2020
Page view(s)
6
Last Week
4
4
Last month
1
1
checked on May 10, 2024
Google ScholarTM
Check
Altmetric
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.