Averaged naive bayes trees: A new extension of aode

Mori Kurokawa; Hiroyuki Yokoyama; Akito Sakurai

doi:10.1007/978-3-642-05224-8_16

Averaged naive bayes trees: A new extension of aode

Mori Kurokawa, Hiroyuki Yokoyama, Akito Sakurai

Department of Industrial and Systems Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

2 Citations (Scopus)

Abstract

Naive Bayes (NB) is a simple Bayesian classifier that assumes the conditional independence and augmented NB (ANB) models are extensions of NB by relaxing the independence assumption. The averaged one-dependence estimators (AODE) is a classifier that averages ODEs, which are ANB models. However, the expressiveness of AODE is still limited by the restricted structure of ODE. In this paper, we propose a model averaging method for NB Trees (NBTs) with flexible structures and present experimental results in terms of classification accuracy. Results of comparative experiments show that our proposed method outperforms AODE on classification accuracy.

Original language	English
Title of host publication	Advances in Machine Learning - First Asian Conference on Machine Learning, ACML 2009, Proceedings
Pages	191-205
Number of pages	15
DOIs	https://doi.org/10.1007/978-3-642-05224-8_16
Publication status	Published - 2009 Dec 1
Event	1st Asian Conference on Machine Learning, ACML 2009 - Nanjing, China Duration: 2009 Nov 2 → 2009 Nov 4

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	5828 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Other

Other	1st Asian Conference on Machine Learning, ACML 2009
Country/Territory	China
City	Nanjing
Period	09/11/2 → 09/11/4

Keywords

Augmented naive Bayes
Averaged one-dependence estimators
Model averaging
Naive Bayes
Naive Bayes trees

ASJC Scopus subject areas

Theoretical Computer Science
Computer Science(all)

Access to Document

10.1007/978-3-642-05224-8_16

Cite this

Kurokawa, M., Yokoyama, H., & Sakurai, A. (2009). Averaged naive bayes trees: A new extension of aode. In Advances in Machine Learning - First Asian Conference on Machine Learning, ACML 2009, Proceedings (pp. 191-205). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5828 LNAI). https://doi.org/10.1007/978-3-642-05224-8_16

Averaged naive bayes trees: A new extension of aode. / Kurokawa, Mori; Yokoyama, Hiroyuki; Sakurai, Akito.
Advances in Machine Learning - First Asian Conference on Machine Learning, ACML 2009, Proceedings. 2009. p. 191-205 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5828 LNAI).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Kurokawa, M, Yokoyama, H & Sakurai, A 2009, Averaged naive bayes trees: A new extension of aode. in Advances in Machine Learning - First Asian Conference on Machine Learning, ACML 2009, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 5828 LNAI, pp. 191-205, 1st Asian Conference on Machine Learning, ACML 2009, Nanjing, China, 09/11/2. https://doi.org/10.1007/978-3-642-05224-8_16

Kurokawa M, Yokoyama H, Sakurai A. Averaged naive bayes trees: A new extension of aode. In Advances in Machine Learning - First Asian Conference on Machine Learning, ACML 2009, Proceedings. 2009. p. 191-205. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-642-05224-8_16

@inproceedings{e99d8e03096948adbec4c7755099ca31,

title = "Averaged naive bayes trees: A new extension of aode",

abstract = "Naive Bayes (NB) is a simple Bayesian classifier that assumes the conditional independence and augmented NB (ANB) models are extensions of NB by relaxing the independence assumption. The averaged one-dependence estimators (AODE) is a classifier that averages ODEs, which are ANB models. However, the expressiveness of AODE is still limited by the restricted structure of ODE. In this paper, we propose a model averaging method for NB Trees (NBTs) with flexible structures and present experimental results in terms of classification accuracy. Results of comparative experiments show that our proposed method outperforms AODE on classification accuracy.",

keywords = "Augmented naive Bayes, Averaged one-dependence estimators, Model averaging, Naive Bayes, Naive Bayes trees",

author = "Mori Kurokawa and Hiroyuki Yokoyama and Akito Sakurai",

year = "2009",

month = dec,

day = "1",

doi = "10.1007/978-3-642-05224-8_16",

language = "English",

isbn = "3642052231",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

pages = "191--205",

booktitle = "Advances in Machine Learning - First Asian Conference on Machine Learning, ACML 2009, Proceedings",

note = "1st Asian Conference on Machine Learning, ACML 2009 ; Conference date: 02-11-2009 Through 04-11-2009",

}

TY - GEN

T1 - Averaged naive bayes trees

T2 - 1st Asian Conference on Machine Learning, ACML 2009

AU - Kurokawa, Mori

AU - Yokoyama, Hiroyuki

AU - Sakurai, Akito

PY - 2009/12/1

Y1 - 2009/12/1

N2 - Naive Bayes (NB) is a simple Bayesian classifier that assumes the conditional independence and augmented NB (ANB) models are extensions of NB by relaxing the independence assumption. The averaged one-dependence estimators (AODE) is a classifier that averages ODEs, which are ANB models. However, the expressiveness of AODE is still limited by the restricted structure of ODE. In this paper, we propose a model averaging method for NB Trees (NBTs) with flexible structures and present experimental results in terms of classification accuracy. Results of comparative experiments show that our proposed method outperforms AODE on classification accuracy.

AB - Naive Bayes (NB) is a simple Bayesian classifier that assumes the conditional independence and augmented NB (ANB) models are extensions of NB by relaxing the independence assumption. The averaged one-dependence estimators (AODE) is a classifier that averages ODEs, which are ANB models. However, the expressiveness of AODE is still limited by the restricted structure of ODE. In this paper, we propose a model averaging method for NB Trees (NBTs) with flexible structures and present experimental results in terms of classification accuracy. Results of comparative experiments show that our proposed method outperforms AODE on classification accuracy.

KW - Augmented naive Bayes

KW - Averaged one-dependence estimators

KW - Model averaging

KW - Naive Bayes

KW - Naive Bayes trees

UR - http://www.scopus.com/inward/record.url?scp=70549098920&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70549098920&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-05224-8_16

DO - 10.1007/978-3-642-05224-8_16

M3 - Conference contribution

AN - SCOPUS:70549098920

SN - 3642052231

SN - 9783642052231

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 191

EP - 205

BT - Advances in Machine Learning - First Asian Conference on Machine Learning, ACML 2009, Proceedings

Y2 - 2 November 2009 through 4 November 2009

ER -

Averaged naive bayes trees: A new extension of aode

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this