Improving supervised learning by feature decomposition

Oded Maimon, Lior Rokach

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

31 Scopus citations


This paper presents the Feature Decomposition Approach for improving supervised learning tasks. While in Feature Selection the aim is to identify a representative set of features from which to construct a classification model, in Feature Decomposition, the goal is to decompose the original set of features into several subsets. A classification model is built for each subset, and then all generated models are combined. This paper presents theoretical and practical aspects of the Feature Decomposition Approach. A greedy procedure, called DOT (Decomposed Oblivious Trees), is developed to decompose the input features set into subsets and to build a classification model for each subset separately. The results achieved in the empirical comparison testing with well-known learning algorithms (like C4.5) indicate the superiority of the feature decomposition approach in learning tasks that contains high number of features and moderate numbers of tuples.

Original languageEnglish
Title of host publicationFoundations of Information and Knowledge Systems - 2nd International Symposium, FoIKS 2002, Proceedings
EditorsThomas Eiter, Klaus-Dieter Schewe
PublisherSpringer Verlag
Number of pages19
ISBN (Print)3540432205, 9783540432203
StatePublished - 2002
Event2nd International Symposium on Foundations of Information and Knowledge Systems, FoIKS 2002 - Salzau Castle, Germany
Duration: 20 Feb 200223 Feb 2002

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Conference2nd International Symposium on Foundations of Information and Knowledge Systems, FoIKS 2002
CitySalzau Castle


Dive into the research topics of 'Improving supervised learning by feature decomposition'. Together they form a unique fingerprint.

Cite this