Presentation 2004-12-17
Succesive Classification by Test Feature Classifier and its Application to Dynamic Recognition problems
Yukinobu SAKATA, Shun'ichi KANEKO, Takayuki TANAKA, Yuji TAKAGI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Fundamental characteristics of successive lerarning by Test Feature Classifier(TFC), which is non-parametric and effective small data, are examined. In the learning, after recognition of a set of unknown objects, they are fed into the classifier in order to obtain a modified classifier. We propose an efficient algorithm for reconstruction of prime test features, which are combinaton feature subsets for getting excellent performance. We apply the proposed successive TFC to dynamic recognition problems from which the characteristic of data changes with progress of time, and examine the characteristic by the experiment which used simulation data.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) classification / test feature classifier / successive learning / dynamic recognition problems
Paper # PRMU2004-138
Date of Issue

Conference Information
Committee PRMU
Conference Date 2004/12/10(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 Pattern Recognition and Media Understanding (PRMU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Succesive Classification by Test Feature Classifier and its Application to Dynamic Recognition problems
Sub Title (in English)
Keyword(1) classification
Keyword(2) test feature classifier
Keyword(3) successive learning
Keyword(4) dynamic recognition problems
1st Author's Name Yukinobu SAKATA
1st Author's Affiliation Graduate School of Graduate School of Information Science and Technology, Hokkaido University()
2nd Author's Name Shun'ichi KANEKO
2nd Author's Affiliation Graduate School of Graduate School of Information Science and Technology, Hokkaido University
3rd Author's Name Takayuki TANAKA
3rd Author's Affiliation Graduate School of Graduate School of Information Science and Technology, Hokkaido University
4th Author's Name Yuji TAKAGI
4th Author's Affiliation PERL, Hitachi Ltd.
Date 2004-12-17
Paper # PRMU2004-138
Volume (vol) vol.104
Number (no) 524
Page pp.pp.-
#Pages 6
Date of Issue