An evidential approach in ensembles

Yaxin Bi, Werner Dubitzky

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we describe an approach to modeling the general process of combining decisions involved in ensembles of classifiers as an evidential reasoning process. This work proposes a novel structure, theoretical properties and manipulation mechanisms for representing classifier decisions as pieces of evidence. The advantage of the representation formalism is that it not only facilitates the distinguishing of trivial focal elements from important ones, resulting in the improvement of the ensemble performance, but it also effectively reduces the computation-time from exponential (as required in the conventional process of combining multiple pieces of evidence) to linear. We have conducted a comparative analysis on the effectiveness of the proposed evidence representation formalism in the text categorization domain. By comparing this method with majority voting and the previous results, we also demonstrate the advantage of this novel approach in combining classifiers.
LanguageEnglish
Title of host publicationUnknown Host Publication
Number of pages6
Publication statusPublished - 2006
EventSAC '06 Proceedings of the 2006 ACM symposium on Applied computing -
Duration: 1 Jan 2006 → …

Conference

ConferenceSAC '06 Proceedings of the 2006 ACM symposium on Applied computing
Period1/01/06 → …

Fingerprint

Classifiers

Cite this

Bi, Y., & Dubitzky, W. (2006). An evidential approach in ensembles. In Unknown Host Publication
Bi, Yaxin ; Dubitzky, Werner. / An evidential approach in ensembles. Unknown Host Publication. 2006.
@inproceedings{5140fd9c4a7c44b29dad307d581957a5,
title = "An evidential approach in ensembles",
abstract = "In this paper, we describe an approach to modeling the general process of combining decisions involved in ensembles of classifiers as an evidential reasoning process. This work proposes a novel structure, theoretical properties and manipulation mechanisms for representing classifier decisions as pieces of evidence. The advantage of the representation formalism is that it not only facilitates the distinguishing of trivial focal elements from important ones, resulting in the improvement of the ensemble performance, but it also effectively reduces the computation-time from exponential (as required in the conventional process of combining multiple pieces of evidence) to linear. We have conducted a comparative analysis on the effectiveness of the proposed evidence representation formalism in the text categorization domain. By comparing this method with majority voting and the previous results, we also demonstrate the advantage of this novel approach in combining classifiers.",
author = "Yaxin Bi and Werner Dubitzky",
year = "2006",
language = "English",
booktitle = "Unknown Host Publication",

}

Bi, Y & Dubitzky, W 2006, An evidential approach in ensembles. in Unknown Host Publication. SAC '06 Proceedings of the 2006 ACM symposium on Applied computing, 1/01/06.

An evidential approach in ensembles. / Bi, Yaxin; Dubitzky, Werner.

Unknown Host Publication. 2006.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - An evidential approach in ensembles

AU - Bi, Yaxin

AU - Dubitzky, Werner

PY - 2006

Y1 - 2006

N2 - In this paper, we describe an approach to modeling the general process of combining decisions involved in ensembles of classifiers as an evidential reasoning process. This work proposes a novel structure, theoretical properties and manipulation mechanisms for representing classifier decisions as pieces of evidence. The advantage of the representation formalism is that it not only facilitates the distinguishing of trivial focal elements from important ones, resulting in the improvement of the ensemble performance, but it also effectively reduces the computation-time from exponential (as required in the conventional process of combining multiple pieces of evidence) to linear. We have conducted a comparative analysis on the effectiveness of the proposed evidence representation formalism in the text categorization domain. By comparing this method with majority voting and the previous results, we also demonstrate the advantage of this novel approach in combining classifiers.

AB - In this paper, we describe an approach to modeling the general process of combining decisions involved in ensembles of classifiers as an evidential reasoning process. This work proposes a novel structure, theoretical properties and manipulation mechanisms for representing classifier decisions as pieces of evidence. The advantage of the representation formalism is that it not only facilitates the distinguishing of trivial focal elements from important ones, resulting in the improvement of the ensemble performance, but it also effectively reduces the computation-time from exponential (as required in the conventional process of combining multiple pieces of evidence) to linear. We have conducted a comparative analysis on the effectiveness of the proposed evidence representation formalism in the text categorization domain. By comparing this method with majority voting and the previous results, we also demonstrate the advantage of this novel approach in combining classifiers.

M3 - Conference contribution

BT - Unknown Host Publication

ER -

Bi Y, Dubitzky W. An evidential approach in ensembles. In Unknown Host Publication. 2006