Analysis of single-objective and multi-objective evolutionary algorithms in keyword cluster optimization

Publikation: Beitrag in Buch/Bericht/TagungsbandKonferenzbeitragBegutachtung

1 Zitat (Scopus)

Abstract

As it is not trivial to cope with the fast growing number of papers published in the field of medicine and biology, intelligent search strategies are needed to be able to access the required information as fast and accurately as possible. In [5] we have proposed a method for keyword clustering as a first step towards an intelligent search strategy in biomedical information retrieval. In this paper we focus on the analysis of the internal dynamics of the evolutionary algorithms applied here using solution encoding specific population diversity analysis, which is also defined in this paper. The population diversity results obtained using evolution strategies, genetic algorithms, genetic algorithms with offspring selection and also a multi-objective approach, the NSGA-II, are discussed here. We see that the diversity of the populations is preserved over the generations, decreasing towards the end of the runs, which indicates a good performance of the selection process.

OriginalspracheEnglisch
TitelComputer Aided Systems Theory, EUROCAST 2011 - 13th International Conference, Revised Selected Papers
Seiten408-415
Seitenumfang8
AuflagePART 1
DOIs
PublikationsstatusVeröffentlicht - 2012
Veranstaltung13th International Conference on Computer Aided Systems Theory, Eurocast 2011 - Las Palmas de Gran Canaria, Spanien
Dauer: 6 Feb 201111 Feb 2011
http://www.iuctc.ulpgc.es/spain/eurocast2011/

Publikationsreihe

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NummerPART 1
Band6927 LNCS
ISSN (Print)0302-9743
ISSN (elektronisch)1611-3349

Konferenz

Konferenz13th International Conference on Computer Aided Systems Theory, Eurocast 2011
Land/GebietSpanien
OrtLas Palmas de Gran Canaria
Zeitraum06.02.201111.02.2011
Internetadresse

Fingerprint

Untersuchen Sie die Forschungsthemen von „Analysis of single-objective and multi-objective evolutionary algorithms in keyword cluster optimization“. Zusammen bilden sie einen einzigartigen Fingerprint.

Zitieren