Quantum inspired automatic clustering algorithms: A comparative study of genetic algorithm and bat algorithm

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27240%2F20%3A10246920" target="_blank" >RIV/61989100:27240/20:10246920 - isvavai.cz</a>

Výsledek na webu

<a href="https://www.degruyter.com/document/doi/10.1515/9783110670707/html" target="_blank" >https://www.degruyter.com/document/doi/10.1515/9783110670707/html</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1515/9783110670707-005" target="_blank" >10.1515/9783110670707-005</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Quantum inspired automatic clustering algorithms: A comparative study of genetic algorithm and bat algorithm

Popis výsledku v původním jazyce

This article is intendant to present two automatic clustering techniques of image datasets, based on quantum inspired framework with two different metaheuristic algorithms, viz., Genetic Algorithm (GA) and Bat Algorithm (BA). This work provides two novel techniques to automatically find out the optimum clusters present in images and also provides a comparative study between the Quantum Inspired Genetic Algorithm (QIGA) and Quantum Inspired Bat Algorithm (QIBA). A comparison is also presented between these quantum inspired algorithms with their analogous classical counterparts. During the experiment, it was perceived that the quantum inspired techniques beat their classical techniques. The comparison was prepared based on the mean values of the fitness, standard deviation, standard error of the computed fitness of the cluster validity index and the optimal computational time. Finally, the supremacy of the algorithms was verified in terms of the p-value which was computed by t-test (statistical superiority test) and ranking of the proposed procedures was produced by the Friedman test. During the computation, the betterment of the fitness was judge by a well-known cluster validity index, named, DB index. The experiments were carried out on four Berkeley image and two real life grey scale images. (C) 2020 Walter de Gruyter GmbH, Berlin/Boston. All rights reserved.

Název v anglickém jazyce

Quantum inspired automatic clustering algorithms: A comparative study of genetic algorithm and bat algorithm

Popis výsledku anglicky

This article is intendant to present two automatic clustering techniques of image datasets, based on quantum inspired framework with two different metaheuristic algorithms, viz., Genetic Algorithm (GA) and Bat Algorithm (BA). This work provides two novel techniques to automatically find out the optimum clusters present in images and also provides a comparative study between the Quantum Inspired Genetic Algorithm (QIGA) and Quantum Inspired Bat Algorithm (QIBA). A comparison is also presented between these quantum inspired algorithms with their analogous classical counterparts. During the experiment, it was perceived that the quantum inspired techniques beat their classical techniques. The comparison was prepared based on the mean values of the fitness, standard deviation, standard error of the computed fitness of the cluster validity index and the optimal computational time. Finally, the supremacy of the algorithms was verified in terms of the p-value which was computed by t-test (statistical superiority test) and ranking of the proposed procedures was produced by the Friedman test. During the computation, the betterment of the fitness was judge by a well-known cluster validity index, named, DB index. The experiments were carried out on four Berkeley image and two real life grey scale images. (C) 2020 Walter de Gruyter GmbH, Berlin/Boston. All rights reserved.

Druh

C - Kapitola v odborné knize

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2020

Kód důvěrnosti údajů

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

Název knihy nebo sborníku

Quantum Machine Learning

ISBN

978-3-11-067070-7

Počet stran výsledku

26

Strana od-do

89-114

Počet stran knihy

131

Název nakladatele

De Gruyter

Místo vydání

Berlin

Kód UT WoS kapitoly

—