Presentation 2004/10/12
ext Classification Based on Ensemble Learning of Document Component Models
Akinori FUJINO, Naonori UEDA, Kazumi SAITO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) For multiclass text classificatin, we propose a new method that considers document components including title, abstract, main content, references, and links. First, a naive Bayes classifier is designed for each document component, in which smoothing parameters are optimally trained by leave-one-out cross validation scheme to boost the generalization performace. Then, based on the maximum entropy principle, a unified classifier is constracted by combined effectively these component classifiers. Through text classification experiments using three sets of real data, we have confirmed the usefulness of the proposed method.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) text classification / ensemble learning / naive Bayes model / maximum entropy principle
Paper # NC2004-80
Date of Issue

Conference Information
Committee NC
Conference Date 2004/10/12(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 JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) ext Classification Based on Ensemble Learning of Document Component Models
Sub Title (in English)
Keyword(1) text classification
Keyword(2) ensemble learning
Keyword(3) naive Bayes model
Keyword(4) maximum entropy principle
1st Author's Name Akinori FUJINO
1st Author's Affiliation NTT Communication Science Laboratories, NTT Corporation()
2nd Author's Name Naonori UEDA
2nd Author's Affiliation NTT Communication Science Laboratories, NTT Corporation
3rd Author's Name Kazumi SAITO
3rd Author's Affiliation NTT Communication Science Laboratories, NTT Corporation
Date 2004/10/12
Paper # NC2004-80
Volume (vol) vol.104
Number (no) 349
Page pp.pp.-
#Pages 6
Date of Issue