Presentation 2006-12-05
Fast Induction of NNC-Trees Based on Dimensionality Reduction
Naoki TOMINAGA, Qiangfu ZHAO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) An NNC-Tree is a decision tree whose non-terminal node contains a nearest neighbor classifier. Compared with the conventional decision trees, the decision rules corresponding to NNC-Trees are more compact, more accurate, and more comprehensible. The induction of NNC-Trees, however, is relatively time-consuming. Thus, to speedup the induction process is an important issue to make NNC-Trees practically useful. To speedup the induction process of NNC-Trees, this paper proposes to reduce the dimensionality of the input space using principal components analysis before induction. Experiments on several public databases show that the proposed method is more effective than the existing algorithm when the number of features is large.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) NNC-Tree / nearest neighbor classifier / principal components analysis / dimensionality reduction / pattern recognition
Paper # NC2006-81
Date of Issue

Conference Information
Committee NC
Conference Date 2006/11/28(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Neurocomputing (NC)
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Fast Induction of NNC-Trees Based on Dimensionality Reduction
Sub Title (in English)
Keyword(1) NNC-Tree
Keyword(2) nearest neighbor classifier
Keyword(3) principal components analysis
Keyword(4) dimensionality reduction
Keyword(5) pattern recognition
1st Author's Name Naoki TOMINAGA
1st Author's Affiliation School of Computer Science and Engineering, The University of Aizu Tsuruga()
2nd Author's Name Qiangfu ZHAO
2nd Author's Affiliation School of Computer Science and Engineering, The University of Aizu Tsuruga
Date 2006-12-05
Paper # NC2006-81
Volume (vol) vol.106
Number (no) 407
Page pp.pp.-
#Pages 6
Date of Issue