Extended Tree Augmented Naive Classifier

Cassio P. de Campos; Marco Cuccu; Giorgio Corani; Marco Zaffalon

doi:10.1007/978-3-319-11433-0_12

Extended Tree Augmented Naive Classifier

Cassio P. de Campos
, Marco Cuccu
, Giorgio Corani
, Marco Zaffalon

School of Electronics, Electrical Engineering and Computer Science

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

9 Citations (Scopus)

Abstract

This work proposes an extended version of the well-known tree-augmented naive Bayes (TAN) classifier where the structure learning step is performed without requiring features to be connected to the class. Based on a modification of Edmonds’ algorithm, our structure learning procedure explores a superset of the structures that are considered by TAN, yet achieves global optimality of the learning score function in a very efficient way (quadratic in the number of features, the same complexity as learning TANs). A range of experiments show that we obtain models with better accuracy than TAN and comparable to the accuracy of the state-of-the-art classifier averaged one-dependence estimator.

Original language	English
Title of host publication	Probabilistic Graphical Models: 7th European Workshop, PGM 2014, Utrecht, The Netherlands, September 17-19, 2014. Proceedings
Editors	Linda C. van der Gaag, Ad J. Feelders
Publisher	Springer Verlag
Pages	176-189
Number of pages	14
Volume	LNAI 8754
ISBN (Electronic)	9783319114330
ISBN (Print)	9783319114323
DOIs	https://doi.org/10.1007/978-3-319-11433-0_12
Publication status	Published - 2015
Event	7th European Workshop, PGM 2014 - Utrecht, Netherlands Duration: 17 Sept 2014 → 19 Sept 2014

Publication series

Name	Lecture Notes in Artificial Intelligence
Volume	LNAI 8794

Conference

Conference	7th European Workshop, PGM 2014
Country/Territory	Netherlands
City	Utrecht
Period	17/09/2014 → 19/09/2014

Bibliographical note

(selected for special issue, blind peer reviewed by >3 reviewers)

Access to Document

10.1007/978-3-319-11433-0_12

http://www.springer.com/us/book/9783319114323

Cite this

de Campos, C. P., Cuccu, M., Corani, G., & Zaffalon, M. (2015). Extended Tree Augmented Naive Classifier. In L. C. van der Gaag, & A. J. Feelders (Eds.), Probabilistic Graphical Models: 7th European Workshop, PGM 2014, Utrecht, The Netherlands, September 17-19, 2014. Proceedings (Vol. LNAI 8754, pp. 176-189). (Lecture Notes in Artificial Intelligence; Vol. LNAI 8794). Springer Verlag. https://doi.org/10.1007/978-3-319-11433-0_12

de Campos, Cassio P. ; Cuccu, Marco ; Corani, Giorgio et al. / Extended Tree Augmented Naive Classifier. Probabilistic Graphical Models: 7th European Workshop, PGM 2014, Utrecht, The Netherlands, September 17-19, 2014. Proceedings. editor / Linda C. van der Gaag ; Ad J. Feelders. Vol. LNAI 8754 Springer Verlag, 2015. pp. 176-189 (Lecture Notes in Artificial Intelligence).

@inproceedings{9fff955dae86471a885071d9fa4f6048,

title = "Extended Tree Augmented Naive Classifier",

abstract = "This work proposes an extended version of the well-known tree-augmented naive Bayes (TAN) classifier where the structure learning step is performed without requiring features to be connected to the class. Based on a modification of Edmonds{\textquoteright} algorithm, our structure learning procedure explores a superset of the structures that are considered by TAN, yet achieves global optimality of the learning score function in a very efficient way (quadratic in the number of features, the same complexity as learning TANs). A range of experiments show that we obtain models with better accuracy than TAN and comparable to the accuracy of the state-of-the-art classifier averaged one-dependence estimator.",

author = "\{de Campos\}, \{Cassio P.\} and Marco Cuccu and Giorgio Corani and Marco Zaffalon",

note = "(selected for special issue, blind peer reviewed by >3 reviewers); 7th European Workshop, PGM 2014 ; Conference date: 17-09-2014 Through 19-09-2014",

year = "2015",

doi = "10.1007/978-3-319-11433-0\_12",

language = "English",

isbn = "9783319114323",

volume = "LNAI 8754",

series = "Lecture Notes in Artificial Intelligence",

publisher = "Springer Verlag",

pages = "176--189",

editor = "\{van der Gaag\}, \{Linda C.\} and Feelders, \{Ad J.\}",

booktitle = "Probabilistic Graphical Models: 7th European Workshop, PGM 2014, Utrecht, The Netherlands, September 17-19, 2014. Proceedings",

address = "Germany",

}

de Campos, CP, Cuccu, M, Corani, G & Zaffalon, M 2015, Extended Tree Augmented Naive Classifier. in LC van der Gaag & AJ Feelders (eds), Probabilistic Graphical Models: 7th European Workshop, PGM 2014, Utrecht, The Netherlands, September 17-19, 2014. Proceedings. vol. LNAI 8754, Lecture Notes in Artificial Intelligence, vol. LNAI 8794, Springer Verlag, pp. 176-189, 7th European Workshop, PGM 2014, Utrecht, Netherlands, 17/09/2014. https://doi.org/10.1007/978-3-319-11433-0_12

Extended Tree Augmented Naive Classifier. / de Campos, Cassio P.; Cuccu, Marco; Corani, Giorgio et al.
Probabilistic Graphical Models: 7th European Workshop, PGM 2014, Utrecht, The Netherlands, September 17-19, 2014. Proceedings. ed. / Linda C. van der Gaag; Ad J. Feelders. Vol. LNAI 8754 Springer Verlag, 2015. p. 176-189 (Lecture Notes in Artificial Intelligence; Vol. LNAI 8794).

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

TY - GEN

T1 - Extended Tree Augmented Naive Classifier

AU - de Campos, Cassio P.

AU - Cuccu, Marco

AU - Corani, Giorgio

AU - Zaffalon, Marco

N1 - (selected for special issue, blind peer reviewed by >3 reviewers)

PY - 2015

Y1 - 2015

N2 - This work proposes an extended version of the well-known tree-augmented naive Bayes (TAN) classifier where the structure learning step is performed without requiring features to be connected to the class. Based on a modification of Edmonds’ algorithm, our structure learning procedure explores a superset of the structures that are considered by TAN, yet achieves global optimality of the learning score function in a very efficient way (quadratic in the number of features, the same complexity as learning TANs). A range of experiments show that we obtain models with better accuracy than TAN and comparable to the accuracy of the state-of-the-art classifier averaged one-dependence estimator.

AB - This work proposes an extended version of the well-known tree-augmented naive Bayes (TAN) classifier where the structure learning step is performed without requiring features to be connected to the class. Based on a modification of Edmonds’ algorithm, our structure learning procedure explores a superset of the structures that are considered by TAN, yet achieves global optimality of the learning score function in a very efficient way (quadratic in the number of features, the same complexity as learning TANs). A range of experiments show that we obtain models with better accuracy than TAN and comparable to the accuracy of the state-of-the-art classifier averaged one-dependence estimator.

U2 - 10.1007/978-3-319-11433-0_12

DO - 10.1007/978-3-319-11433-0_12

M3 - Conference contribution

SN - 9783319114323

VL - LNAI 8754

T3 - Lecture Notes in Artificial Intelligence

SP - 176

EP - 189

BT - Probabilistic Graphical Models: 7th European Workshop, PGM 2014, Utrecht, The Netherlands, September 17-19, 2014. Proceedings

A2 - van der Gaag, Linda C.

A2 - Feelders, Ad J.

PB - Springer Verlag

T2 - 7th European Workshop, PGM 2014

Y2 - 17 September 2014 through 19 September 2014

ER -

de Campos CP, Cuccu M, Corani G, Zaffalon M. Extended Tree Augmented Naive Classifier. In van der Gaag LC, Feelders AJ, editors, Probabilistic Graphical Models: 7th European Workshop, PGM 2014, Utrecht, The Netherlands, September 17-19, 2014. Proceedings. Vol. LNAI 8754. Springer Verlag. 2015. p. 176-189. (Lecture Notes in Artificial Intelligence). doi: 10.1007/978-3-319-11433-0_12

Extended Tree Augmented Naive Classifier

Abstract

Publication series

Conference

Bibliographical note

Access to Document

Fingerprint

Cite this