Constructive meta-learning with machine learning method repositories

Hidenao Abe, Takahira Yamaguchi

Research output: Contribution to journalConference articlepeer-review

22 Citations (Scopus)


Here is discussed what is constructive meta-learning and how it goes well compared with selective meta-learning that already becomes popular. Selective meta-learning takes multiple learning schemes with the following different ways: bagging, boosting, cascading and stacking methods. On the other hand, constructive meta-learning constructs the learning scheme proper to a given data set. We have implemented constructive meta-learning by recomposing methods into learning schemes with mining (inductive learning) method repositories that come from decomposition of popular mining algorithms. To evaluate our constructive meta-learning, we have done the comparison of the performances of our constructive meta-learning and those of two stacking methods, using UCI/ML common data sets. It has shown us that our constructive meta-learning goes better than the two stacking methods. Furthermore, it turns out to be promising that we apply constructive meta-learning to meta-learner in selective meta-learning.

Original languageEnglish
Pages (from-to)502-511
Number of pages10
JournalLecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science)
Publication statusPublished - 2004
Externally publishedYes
Event17th International Conference on Industrial and Engineering Applications of Artificial Intelligence and Expert Systems, IEA/AIE 2004 - Ottowa, Ont., Canada
Duration: 2004 May 172004 May 20

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science


Dive into the research topics of 'Constructive meta-learning with machine learning method repositories'. Together they form a unique fingerprint.

Cite this