Matrix constructions of centroid sets for classification systems

Morshed Chowdhury, Jemal Abawajy, Herbert Jelinek, Andrei Kelarev, Joe Ryan

Research output: Contribution to journalArticle

3 Downloads (Pure)

Abstract

t. This article continues the investigation of matrix constructions motivated by their applications to the design of classification systems. Our main theorems strengthen and generalize previous results by describing all centroid sets for classification systems that can be generated as one-sided ideals with the largest weight in structural matrix semirings. Centroid sets are well known in data mining, where they are used for the design of centroid-based classification systems, as well as for the design of multiple classification systems combining several individual classifiers.
Original languageEnglish
Pages (from-to)2397-2403
Number of pages7
JournalFilomat
Volume30
Issue number9
DOIs
Publication statusPublished - 2016

Fingerprint

Centroid
Semiring
Data Mining
Continue
Classifier
Generalise
Theorem
Design

Cite this

Chowdhury, M., Abawajy, J., Jelinek, H., Kelarev, A., & Ryan, J. (2016). Matrix constructions of centroid sets for classification systems. Filomat, 30(9), 2397-2403. https://doi.org/10.2298/FIL1609397C
Chowdhury, Morshed ; Abawajy, Jemal ; Jelinek, Herbert ; Kelarev, Andrei ; Ryan, Joe. / Matrix constructions of centroid sets for classification systems. In: Filomat. 2016 ; Vol. 30, No. 9. pp. 2397-2403.
@article{e2cf0389349042e8b6e5442f62028a8f,
title = "Matrix constructions of centroid sets for classification systems",
abstract = "t. This article continues the investigation of matrix constructions motivated by their applications to the design of classification systems. Our main theorems strengthen and generalize previous results by describing all centroid sets for classification systems that can be generated as one-sided ideals with the largest weight in structural matrix semirings. Centroid sets are well known in data mining, where they are used for the design of centroid-based classification systems, as well as for the design of multiple classification systems combining several individual classifiers.",
keywords = "Open access version available",
author = "Morshed Chowdhury and Jemal Abawajy and Herbert Jelinek and Andrei Kelarev and Joe Ryan",
note = "Imported on 12 Apr 2017 - DigiTool details were: Journal title (773t) = FILOMAT. ISSNs: 0354-5180;",
year = "2016",
doi = "10.2298/FIL1609397C",
language = "English",
volume = "30",
pages = "2397--2403",
journal = "Filomat",
issn = "0354-5180",
publisher = "Universitet of Nis",
number = "9",

}

Chowdhury, M, Abawajy, J, Jelinek, H, Kelarev, A & Ryan, J 2016, 'Matrix constructions of centroid sets for classification systems', Filomat, vol. 30, no. 9, pp. 2397-2403. https://doi.org/10.2298/FIL1609397C

Matrix constructions of centroid sets for classification systems. / Chowdhury, Morshed; Abawajy, Jemal; Jelinek, Herbert; Kelarev, Andrei; Ryan, Joe.

In: Filomat, Vol. 30, No. 9, 2016, p. 2397-2403.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Matrix constructions of centroid sets for classification systems

AU - Chowdhury, Morshed

AU - Abawajy, Jemal

AU - Jelinek, Herbert

AU - Kelarev, Andrei

AU - Ryan, Joe

N1 - Imported on 12 Apr 2017 - DigiTool details were: Journal title (773t) = FILOMAT. ISSNs: 0354-5180;

PY - 2016

Y1 - 2016

N2 - t. This article continues the investigation of matrix constructions motivated by their applications to the design of classification systems. Our main theorems strengthen and generalize previous results by describing all centroid sets for classification systems that can be generated as one-sided ideals with the largest weight in structural matrix semirings. Centroid sets are well known in data mining, where they are used for the design of centroid-based classification systems, as well as for the design of multiple classification systems combining several individual classifiers.

AB - t. This article continues the investigation of matrix constructions motivated by their applications to the design of classification systems. Our main theorems strengthen and generalize previous results by describing all centroid sets for classification systems that can be generated as one-sided ideals with the largest weight in structural matrix semirings. Centroid sets are well known in data mining, where they are used for the design of centroid-based classification systems, as well as for the design of multiple classification systems combining several individual classifiers.

KW - Open access version available

U2 - 10.2298/FIL1609397C

DO - 10.2298/FIL1609397C

M3 - Article

VL - 30

SP - 2397

EP - 2403

JO - Filomat

JF - Filomat

SN - 0354-5180

IS - 9

ER -