Aggregation Functions in Flexible Classification by Ordinal Sums

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27510%2F22%3A10250618" target="_blank" >RIV/61989100:27510/22:10250618 - isvavai.cz</a>

Výsledek na webu

<a href="https://ipmu2022.disco.unimib.it/proceedings/" target="_blank" >https://ipmu2022.disco.unimib.it/proceedings/</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-031-08971-8_31" target="_blank" >10.1007/978-3-031-08971-8_31</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Aggregation Functions in Flexible Classification by Ordinal Sums

Popis výsledku v původním jazyce

The structure of ordinal sums of conjunctive and disjunctive functions is convenient for classification into the classes Yes, No and Maybe containing a tendency to the classes Yes and No. It especially holds when task is expressed by short vague requirements. The averaging part is covered by any averaging function. Concerning this part (class Maybe), functions having annihilator 0 usually are not suitable due to A(1, 0 ) = A(0, 1 ) = 0. The dual observation holds for functions having annihilator equal to 1. The classification by uninorms has reached to the same conclusion. This work proposes a parametric class of quasi-arithmetic means with the convex combination of the geometric mean and its dual geometric mean. Regarding classification into classes Yes and No, the parametrized family of nilpotent t-norms and t-conorms is proposed. This consideration creates the frame for learning function&apos;s parameter from the user inputs and labelled data. This research activity also contributes to the field of explainable computational intelligence. The results are supported by illustrative example. Finally, the discussion and the future research activities conclude the paper. (C) 2022, Springer Nature Switzerland AG.

Název v anglickém jazyce

Aggregation Functions in Flexible Classification by Ordinal Sums

Popis výsledku anglicky

The structure of ordinal sums of conjunctive and disjunctive functions is convenient for classification into the classes Yes, No and Maybe containing a tendency to the classes Yes and No. It especially holds when task is expressed by short vague requirements. The averaging part is covered by any averaging function. Concerning this part (class Maybe), functions having annihilator 0 usually are not suitable due to A(1, 0 ) = A(0, 1 ) = 0. The dual observation holds for functions having annihilator equal to 1. The classification by uninorms has reached to the same conclusion. This work proposes a parametric class of quasi-arithmetic means with the convex combination of the geometric mean and its dual geometric mean. Regarding classification into classes Yes and No, the parametrized family of nilpotent t-norms and t-conorms is proposed. This consideration creates the frame for learning function&apos;s parameter from the user inputs and labelled data. This research activity also contributes to the field of explainable computational intelligence. The results are supported by illustrative example. Finally, the discussion and the future research activities conclude the paper. (C) 2022, Springer Nature Switzerland AG.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10200 - Computer and information sciences

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2022

Kód důvěrnosti údajů

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

Název statě ve sborníku

Communications in Computer and Information Science. Volume 1601

ISBN

978-3-031-08970-1

ISSN

1865-0929

e-ISSN

1865-0937

Počet stran výsledku

12

Strana od-do

372-383

Název nakladatele

Springer

Místo vydání

Cham

Místo konání akce

Milán

Datum konání akce

11. 7. 2022

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—