Handle URI:
http://hdl.handle.net/10754/597845
Title:
Context-sensitive intra-class clustering
Authors:
Yu, Yingwei; Gutierrez-Osuna, Ricardo; Choe, Yoonsuck
Abstract:
This paper describes a new semi-supervised learning algorithm for intra-class clustering (ICC). ICC partitions each class into sub-classes in order to minimize overlap across clusters from different classes. This is achieved by allowing partitioning of a certain class to be assisted by data points from other classes in a context-dependent fashion. The result is that overlap across sub-classes (both within- and across class) is greatly reduced. ICC is particularly useful when combined with algorithms that assume that each class has a unimodal Gaussian distribution (e.g., Linear Discriminant Analysis (LDA), quadratic classifiers), an assumption that is not always true in many real-world situations. ICC can help partition non-Gaussian, multimodal distributions to overcome such a problem. In this sense, ICC works as a preprocessor. Experiments with our ICC algorithm on synthetic data sets and real-world data sets indicated that it can significantly improve the performance of LDA and quadratic classifiers. We expect our approach to be applicable to a broader class of pattern recognition problems where class-conditional densities are significantly non-Gaussian or multi-modal. © 2013 Elsevier Ltd. All rights reserved.
Citation:
Yu Y, Gutierrez-Osuna R, Choe Y (2014) Context-sensitive intra-class clustering. Pattern Recognition Letters 37: 85–93. Available: http://dx.doi.org/10.1016/j.patrec.2013.04.031.
Publisher:
Elsevier BV
Journal:
Pattern Recognition Letters
KAUST Grant Number:
KUSC1-016-04
Issue Date:
Feb-2014
DOI:
10.1016/j.patrec.2013.04.031
Type:
Article
ISSN:
0167-8655
Sponsors:
This publication is based in part on work supported by Award No. KUSC1-016-04, made by King Abdullah University of Science and Technology (KAUST).
Appears in Collections:
Publications Acknowledging KAUST Support

Full metadata record

DC FieldValue Language
dc.contributor.authorYu, Yingweien
dc.contributor.authorGutierrez-Osuna, Ricardoen
dc.contributor.authorChoe, Yoonsucken
dc.date.accessioned2016-02-25T12:57:41Zen
dc.date.available2016-02-25T12:57:41Zen
dc.date.issued2014-02en
dc.identifier.citationYu Y, Gutierrez-Osuna R, Choe Y (2014) Context-sensitive intra-class clustering. Pattern Recognition Letters 37: 85–93. Available: http://dx.doi.org/10.1016/j.patrec.2013.04.031.en
dc.identifier.issn0167-8655en
dc.identifier.doi10.1016/j.patrec.2013.04.031en
dc.identifier.urihttp://hdl.handle.net/10754/597845en
dc.description.abstractThis paper describes a new semi-supervised learning algorithm for intra-class clustering (ICC). ICC partitions each class into sub-classes in order to minimize overlap across clusters from different classes. This is achieved by allowing partitioning of a certain class to be assisted by data points from other classes in a context-dependent fashion. The result is that overlap across sub-classes (both within- and across class) is greatly reduced. ICC is particularly useful when combined with algorithms that assume that each class has a unimodal Gaussian distribution (e.g., Linear Discriminant Analysis (LDA), quadratic classifiers), an assumption that is not always true in many real-world situations. ICC can help partition non-Gaussian, multimodal distributions to overcome such a problem. In this sense, ICC works as a preprocessor. Experiments with our ICC algorithm on synthetic data sets and real-world data sets indicated that it can significantly improve the performance of LDA and quadratic classifiers. We expect our approach to be applicable to a broader class of pattern recognition problems where class-conditional densities are significantly non-Gaussian or multi-modal. © 2013 Elsevier Ltd. All rights reserved.en
dc.description.sponsorshipThis publication is based in part on work supported by Award No. KUSC1-016-04, made by King Abdullah University of Science and Technology (KAUST).en
dc.publisherElsevier BVen
dc.subjectClusteringen
dc.subjectICCen
dc.subjectIntra-class clusteringen
dc.subjectLDAen
dc.subjectLinear discriminant analysisen
dc.subjectSemi-supervised learningen
dc.titleContext-sensitive intra-class clusteringen
dc.typeArticleen
dc.identifier.journalPattern Recognition Lettersen
dc.contributor.institutionIHS Inc., Houston, United Statesen
dc.contributor.institutionTexas A and M University, College Station, United Statesen
kaust.grant.numberKUSC1-016-04en
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.