A methodology for identifying interesting association rules by combining objective and subjective measures

Fuente: Redalyc
Salvato in:
Dettagli Bibliografici
Autore principale: Roberta Akemi Sinoara
Natura: Artículo científico
Lingua:en
Pubblicazione: Asociación Española para la Inteligencia Artificial 2006
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1876426019569664000
author Roberta Akemi Sinoara
author_facet Roberta Akemi Sinoara
contents A methodology for identifying interesting association rules by combining objective and subjective measures Roberta Akemi Sinoara Solange Oliveira Rezende Ingeniería Data Mining Association Rules Evaluation Measures Evaluation measures, objective and subjective, are used to assist users in finding interesting association rules.Objective measures are more general, but they can be insufficient because they do not consider user’s anddomain features. However, getting user’s knowledge and interest needed to calculate subjective measures canbe a difficult task. Thus, this work presents a methodology to identify interesting association rules combininganalysis with objective and subjective measures. This methodology aims to use the advantages of each kindof measure and to make user’s participation easier. Objective measures are used to select some potentiallyinteresting rules for the user’s evaluation. Through this evaluation the user’s subjectivity is obtained andused to calculate the subjective measures. Then, the subjective measures assist in identifying the interestingrules. In order to exemplify the methodology application, an experiment was carried out with a real databaseand the methodology showed to be feasible. 2006 artículo científico 1137-3601 https://www.redalyc.org/articulo.oa?id=92503204 en http://www.redalyc.org/revista.oa?id=925 Inteligencia Artificial. Revista Iberoamericana de Inteligencia Artificial application/pdf Asociación Española para la Inteligencia Artificial Inteligencia Artificial. Revista Iberoamericana de Inteligencia Artificial (España) Num.32 Vol.10
format Artículo científico
id redalyc_92503204
institution Redalyc
language en
publishDate 2006
publisher Asociación Española para la Inteligencia Artificial
spellingShingle A methodology for identifying interesting association rules by combining objective and subjective measures
Roberta Akemi Sinoara
Ingeniería
Data Mining
Association Rules
Evaluation Measures
A methodology for identifying interesting association rules by combining objective and subjective measures Roberta Akemi Sinoara Solange Oliveira Rezende Ingeniería Data Mining Association Rules Evaluation Measures Evaluation measures, objective and subjective, are used to assist users in finding interesting association rules.Objective measures are more general, but they can be insufficient because they do not consider user’s anddomain features. However, getting user’s knowledge and interest needed to calculate subjective measures canbe a difficult task. Thus, this work presents a methodology to identify interesting association rules combininganalysis with objective and subjective measures. This methodology aims to use the advantages of each kindof measure and to make user’s participation easier. Objective measures are used to select some potentiallyinteresting rules for the user’s evaluation. Through this evaluation the user’s subjectivity is obtained andused to calculate the subjective measures. Then, the subjective measures assist in identifying the interestingrules. In order to exemplify the methodology application, an experiment was carried out with a real databaseand the methodology showed to be feasible. 2006 artículo científico 1137-3601 https://www.redalyc.org/articulo.oa?id=92503204 en http://www.redalyc.org/revista.oa?id=925 Inteligencia Artificial. Revista Iberoamericana de Inteligencia Artificial application/pdf Asociación Española para la Inteligencia Artificial Inteligencia Artificial. Revista Iberoamericana de Inteligencia Artificial (España) Num.32 Vol.10
title A methodology for identifying interesting association rules by combining objective and subjective measures
topic Ingeniería
Data Mining
Association Rules
Evaluation Measures
url https://www.redalyc.org/articulo.oa?id=92503204