Logotipo do repositório
 

Publicação:
Dimensionality reduction for the algorithm recommendation problem

dc.contributor.authorAlcobaca, Edesio
dc.contributor.authorMantovani, Rafael G.
dc.contributor.authorRossi, Andre L. D. [UNESP]
dc.contributor.authorDe Carvalho, André C.P.L.F.
dc.contributor.institutionUniversidade de São Paulo (USP)
dc.contributor.institutionUniversidade Estadual Paulista (Unesp)
dc.date.accessioned2019-10-06T16:15:51Z
dc.date.available2019-10-06T16:15:51Z
dc.date.issued2018-12-13
dc.description.abstractGiven the increase in data generation, as many algorithms have become available in recent years, the algorithm recommendation problem has attracted increasing attention in Machine Learning. This problem has been addressed in the Machine Learning community as a learning task at the meta-level where the most suitable algorithm has to be recommended for a specific dataset. Since it is not trivial to define which characteristics are the most useful for a specific domain, several meta-features have been proposed and used, increasing the meta-data meta-feature dimension. This study investigates the influence of dimensionality reduction techniques on the quality of the algorithm recommendation process. Experiments were carried out with 15 algorithm recommendation problems from the Aslib library, 4 meta-learners, and 3 dimensionality reduction techniques. The experimental results showed that linear aggregation techniques, such as PCA and LDA, can be used in algorithm recommendation problems to reduce the number of meta-features and computational cost without losing predictive performance.en
dc.description.affiliationInstitute of Mathematics and Computer Sciences University of Saõ Paulo (ICMC/USP)
dc.description.affiliationSaõ Paulo State University (UNESP) Campus of Itapeva
dc.description.affiliationUnespSaõ Paulo State University (UNESP) Campus of Itapeva
dc.format.extent318-323
dc.identifierhttp://dx.doi.org/10.1109/BRACIS.2018.00062
dc.identifier.citationProceedings - 2018 Brazilian Conference on Intelligent Systems, BRACIS 2018, p. 318-323.
dc.identifier.doi10.1109/BRACIS.2018.00062
dc.identifier.scopus2-s2.0-85060862209
dc.identifier.urihttp://hdl.handle.net/11449/188682
dc.language.isoeng
dc.relation.ispartofProceedings - 2018 Brazilian Conference on Intelligent Systems, BRACIS 2018
dc.rights.accessRightsAcesso aberto
dc.sourceScopus
dc.subjectAlgorithm recommendation
dc.subjectDimensionality reduction
dc.subjectMeta-learning
dc.titleDimensionality reduction for the algorithm recommendation problemen
dc.typeTrabalho apresentado em evento
dspace.entity.typePublication
unesp.author.lattes5604829226181486[3]
unesp.author.orcid0000-0001-6388-7479[3]
unesp.campusUniversidade Estadual Paulista (UNESP), Instituto de Ciências e Engenharia, Itapevapt
unesp.departmentEngenharia Industrial Madeireira - ICEpt

Arquivos