Parallel model refinement of inductive applications using method repository

Hidenao Abe, Takahira Yamaguchi

Research output: Contribution to journalArticle

Abstract

Here is presented parallel CAMLET that is a platform for automatic composition of inductive applications with method repositories that organize many inductive learning methods. After having implemented CAMLET on a UNIX parallel machine with Perl and C languages, we have done the case studies of constructing inductive applications for eight different data sets from the StatLog repository. To find out an efficient search method, we have done the following experiments: a random search, a GA based search, and two hybrid searches with unifying each global search and the local search which uses meta-rules for refining a specification. That have shown us that the hybrid search works better than the other search methods. Furthermore, comparing the accuracy of inductive applications composed by parallel CMALET with that of popular twenty four inductive algorithms, we have shown that parallel CAMLET support a user in doing model selection in KDD engineering processes.

Original languageEnglish
Pages (from-to)647-657
Number of pages11
JournalTransactions of the Japanese Society for Artificial Intelligence
Volume17
Issue number5
DOIs
Publication statusPublished - 2002
Externally publishedYes

Fingerprint

UNIX
Refining
Specifications
Chemical analysis
Experiments

Keywords

  • Automatic composition
  • Inductive applications
  • Meta-learning
  • Repository

ASJC Scopus subject areas

  • Artificial Intelligence

Cite this

Parallel model refinement of inductive applications using method repository. / Abe, Hidenao; Yamaguchi, Takahira.

In: Transactions of the Japanese Society for Artificial Intelligence, Vol. 17, No. 5, 2002, p. 647-657.

Research output: Contribution to journalArticle

@article{79f908cac2dd47a8819734a6acba89b5,
title = "Parallel model refinement of inductive applications using method repository",
abstract = "Here is presented parallel CAMLET that is a platform for automatic composition of inductive applications with method repositories that organize many inductive learning methods. After having implemented CAMLET on a UNIX parallel machine with Perl and C languages, we have done the case studies of constructing inductive applications for eight different data sets from the StatLog repository. To find out an efficient search method, we have done the following experiments: a random search, a GA based search, and two hybrid searches with unifying each global search and the local search which uses meta-rules for refining a specification. That have shown us that the hybrid search works better than the other search methods. Furthermore, comparing the accuracy of inductive applications composed by parallel CMALET with that of popular twenty four inductive algorithms, we have shown that parallel CAMLET support a user in doing model selection in KDD engineering processes.",
keywords = "Automatic composition, Inductive applications, Meta-learning, Repository",
author = "Hidenao Abe and Takahira Yamaguchi",
year = "2002",
doi = "10.1527/tjsai.17.647",
language = "English",
volume = "17",
pages = "647--657",
journal = "Transactions of the Japanese Society for Artificial Intelligence",
issn = "1346-0714",
publisher = "Japanese Society for Artificial Intelligence",
number = "5",

}

TY - JOUR

T1 - Parallel model refinement of inductive applications using method repository

AU - Abe, Hidenao

AU - Yamaguchi, Takahira

PY - 2002

Y1 - 2002

N2 - Here is presented parallel CAMLET that is a platform for automatic composition of inductive applications with method repositories that organize many inductive learning methods. After having implemented CAMLET on a UNIX parallel machine with Perl and C languages, we have done the case studies of constructing inductive applications for eight different data sets from the StatLog repository. To find out an efficient search method, we have done the following experiments: a random search, a GA based search, and two hybrid searches with unifying each global search and the local search which uses meta-rules for refining a specification. That have shown us that the hybrid search works better than the other search methods. Furthermore, comparing the accuracy of inductive applications composed by parallel CMALET with that of popular twenty four inductive algorithms, we have shown that parallel CAMLET support a user in doing model selection in KDD engineering processes.

AB - Here is presented parallel CAMLET that is a platform for automatic composition of inductive applications with method repositories that organize many inductive learning methods. After having implemented CAMLET on a UNIX parallel machine with Perl and C languages, we have done the case studies of constructing inductive applications for eight different data sets from the StatLog repository. To find out an efficient search method, we have done the following experiments: a random search, a GA based search, and two hybrid searches with unifying each global search and the local search which uses meta-rules for refining a specification. That have shown us that the hybrid search works better than the other search methods. Furthermore, comparing the accuracy of inductive applications composed by parallel CMALET with that of popular twenty four inductive algorithms, we have shown that parallel CAMLET support a user in doing model selection in KDD engineering processes.

KW - Automatic composition

KW - Inductive applications

KW - Meta-learning

KW - Repository

UR - http://www.scopus.com/inward/record.url?scp=18444408693&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=18444408693&partnerID=8YFLogxK

U2 - 10.1527/tjsai.17.647

DO - 10.1527/tjsai.17.647

M3 - Article

AN - SCOPUS:18444408693

VL - 17

SP - 647

EP - 657

JO - Transactions of the Japanese Society for Artificial Intelligence

JF - Transactions of the Japanese Society for Artificial Intelligence

SN - 1346-0714

IS - 5

ER -