DSpace

Future University Hakodate Academic Archive >
Faculty and Students >
Dept. of Media Architecture >
Pitoyo, Hartono >

Please use this identifier to cite or link to this item: http://hdl.handle.net/10445/5283

Title: Ensemble of Linear Experts as an Interpretable Piecewise Linear Classifier
Authors: Hartono, Pitoyo
Abstract: In this study we propose a new ensemble model composed of several linear perceptrons. The objective of this study is to build a piecewise-linear classifier that is not only competitive to Multilayer Perceptrons(MLP) in generalization performance but also interpretable in the form of human-comprehensible rules. We propose a simple competitive training method that allows the ensemble to effectively divide a given training space into several sub-spaces based on so called "confidence value", and train each module to obtain a linear rule within the allocated sub-space. The linearity of the ensemble's module significantly simplifies the rule extraction process.
Research Achievement Classification: 原著論文/Original Paper
Type: Journal Article
Peer Review: あり/yes
Solo/Joint Author(s): 単著/solo
Published journal or presented
academic conference: 
Innovative Computing, Information and Control Express Letter 
Volume: 2
Number: 3
Spage: 295
Epage: 303
Date: 2008
Appears in Collections:Pitoyo, Hartono

Files in This Item:

There are no files associated with this item.

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

 

Copyright © 2010-2012 FUTURE UNIVERSITY HAKODATE.
DSpace Software Copyright © 2002-2010  Duraspace