Testing of cluster algorithms for invariance with respect to numbering of objects

I. Z. Batyrshin; R. F. Khabibulin

Testing of cluster algorithms for invariance with respect to numbering of objects

Producción científica: Contribución a una revista › Artículo › revisión exhaustiva

Resumen

A new approach to testing cluster algorithms for invariance with respect to numbering of objects is described. The approach is based on the algorithm testing by symmetric data. This approach makes it possible to avoid multiple counting or transformation of the similarity matrix and requires only single application of the algorithm.

Idioma original	Inglés
Páginas (desde-hasta)	165-168
Número de páginas	4
Publicación	Izvestiya Akademii Nauk. Teoriya i Sistemy Upravleniya
N.º	2
Estado	Publicada - mar. 1997
Publicado de forma externa	Sí

Otros archivos y enlaces

Enlace a la publicación en Scopus

Citar esto

@article{a327cadef383405bb8f80ece7c592f1a,

title = "Testing of cluster algorithms for invariance with respect to numbering of objects",

abstract = "A new approach to testing cluster algorithms for invariance with respect to numbering of objects is described. The approach is based on the algorithm testing by symmetric data. This approach makes it possible to avoid multiple counting or transformation of the similarity matrix and requires only single application of the algorithm.",

author = "Batyrshin, {I. Z.} and Khabibulin, {R. F.}",

year = "1997",

month = mar,

language = "Ingl{\'e}s",

pages = "165--168",

journal = "Izvestiya Akademii Nauk. Teoriya i Sistemy Upravleniya",

issn = "0002-3388",

publisher = "Izdatel'stvo Nauka",

number = "2",

}

TY - JOUR

T1 - Testing of cluster algorithms for invariance with respect to numbering of objects

AU - Batyrshin, I. Z.

AU - Khabibulin, R. F.

PY - 1997/3

Y1 - 1997/3

N2 - A new approach to testing cluster algorithms for invariance with respect to numbering of objects is described. The approach is based on the algorithm testing by symmetric data. This approach makes it possible to avoid multiple counting or transformation of the similarity matrix and requires only single application of the algorithm.

AB - A new approach to testing cluster algorithms for invariance with respect to numbering of objects is described. The approach is based on the algorithm testing by symmetric data. This approach makes it possible to avoid multiple counting or transformation of the similarity matrix and requires only single application of the algorithm.

UR - http://www.scopus.com/inward/record.url?scp=0031091685&partnerID=8YFLogxK

M3 - Artículo

SN - 0002-3388

SP - 165

EP - 168

JO - Izvestiya Akademii Nauk. Teoriya i Sistemy Upravleniya

JF - Izvestiya Akademii Nauk. Teoriya i Sistemy Upravleniya

IS - 2

ER -

Testing of cluster algorithms for invariance with respect to numbering of objects

Resumen

Otros archivos y enlaces

Huella

Citar esto