A Generalized Kernel Approach to Dissimilarity Based Classification

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21260%2F01%3A06071295" target="_blank" >RIV/68407700:21260/01:06071295 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

A Generalized Kernel Approach to Dissimilarity Based Classification

Popis výsledku v původním jazyce

Usually, objects to be classified are represented by features. Ing this paper, we discuss an alternative object representation based on dissimilarity values. If such distances separate the classes well, the nearest neighbor method offers from its sensitivity to noisy examples. We show that other, more global classification techniques are preferable to the nearest neighbor rule, in such cases. For classification purposes, two different ways of using generalized dissimilarity embedded in a pseudo-Euclidean space and the classification task is performed there. In the second approach, classifiers are built directly on distance kernels. Both approaches are described theoretically and then compared using experiments with different dissimilarity measures andthen compared using experiments with different dissimilarity measures and datasets including degraded data simulating the problem of missing values.

Název v anglickém jazyce

A Generalized Kernel Approach to Dissimilarity Based Classification

Popis výsledku anglicky

Usually, objects to be classified are represented by features. Ing this paper, we discuss an alternative object representation based on dissimilarity values. If such distances separate the classes well, the nearest neighbor method offers from its sensitivity to noisy examples. We show that other, more global classification techniques are preferable to the nearest neighbor rule, in such cases. For classification purposes, two different ways of using generalized dissimilarity embedded in a pseudo-Euclidean space and the classification task is performed there. In the second approach, classifiers are built directly on distance kernels. Both approaches are described theoretically and then compared using experiments with different dissimilarity measures andthen compared using experiments with different dissimilarity measures and datasets including degraded data simulating the problem of missing values.

Druh

J_x - Nezařazeno - Článek v odborném periodiku (Jimp, Jsc a Jost)

CEP obor

BA - Obecná matematika

OECD FORD obor

—

Projekt

—

Návaznosti

Z - Vyzkumny zamer (s odkazem do CEZ)

Rok uplatnění

2001

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

Machine Learning Research on Kernel Methods

ISSN

1533-7928

e-ISSN

—

Svazek periodika

2

Číslo periodika v rámci svazku

2

Stát vydavatele periodika

US - Spojené státy americké

Počet stran výsledku

37

Strana od-do

175-211

Kód UT WoS článku

—

EID výsledku v databázi Scopus

—