Learning naive Bayes models for multiple-instance learning with label proportions

Jerónimo Hernández-González; Iñaki Inza

doi:10.1007/978-3-642-25274-7_14

Learning naive Bayes models for multiple-instance learning with label proportions

Jerónimo Hernández-González
, Iñaki Inza

University of the Basque Country

Research output: Chapter in Book/Conference proceeding › Conference proceeding › peer-review

6 Citations (Scopus)

Abstract

This paper deals with the problem of multi-instance learning when label proportions are provided. In this classification problem, the instances of the dataset are divided into disjoint groups, where there is no certainty about the labels associated with individual samples. However, in each group the number of instances that belong to each class is known. We propose several versions of an EM-algorithm that learns naive Bayes models to deal with the exposed problem. The proposed algorithms are evaluated on synthetic and real datasets, and compared with state-of-the-art approaches. The obtained results show a competitive behaviour of our proposals.

Original language	English
Title of host publication	Advances in Artificial Intelligence - 14th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2011, Proceedings
Publisher	Springer Verlag
Pages	134-144
Number of pages	11
ISBN (Print)	9783642252730
DOIs	https://doi.org/10.1007/978-3-642-25274-7_14
Publication status	Published - 2011
Externally published	Yes
Event	14th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2011 - La Laguna, Spain Duration: 7 Nov 2011 → 11 Nov 2011

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	7023 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	14th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2011
Country/Territory	Spain
City	La Laguna
Period	7/11/11 → 11/11/11

Keywords

EM algorithm
Multiple-instance learning with label proportions
Naive Bayes
supervised classification

Access to Document

10.1007/978-3-642-25274-7_14

Cite this

Hernández-González, J., & Inza, I. (2011). Learning naive Bayes models for multiple-instance learning with label proportions. In Advances in Artificial Intelligence - 14th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2011, Proceedings (pp. 134-144). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7023 LNAI). Springer Verlag. https://doi.org/10.1007/978-3-642-25274-7_14

Hernández-González, Jerónimo ; Inza, Iñaki. / Learning naive Bayes models for multiple-instance learning with label proportions. Advances in Artificial Intelligence - 14th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2011, Proceedings. Springer Verlag, 2011. pp. 134-144 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{3645b4f1f9e64bc5a281a2caeff9b623,

title = "Learning naive Bayes models for multiple-instance learning with label proportions",

abstract = "This paper deals with the problem of multi-instance learning when label proportions are provided. In this classification problem, the instances of the dataset are divided into disjoint groups, where there is no certainty about the labels associated with individual samples. However, in each group the number of instances that belong to each class is known. We propose several versions of an EM-algorithm that learns naive Bayes models to deal with the exposed problem. The proposed algorithms are evaluated on synthetic and real datasets, and compared with state-of-the-art approaches. The obtained results show a competitive behaviour of our proposals.",

keywords = "EM algorithm, Multiple-instance learning with label proportions, Naive Bayes, supervised classification",

author = "Jer{\'o}nimo Hern{\'a}ndez-Gonz{\'a}lez and I{\~n}aki Inza",

year = "2011",

doi = "10.1007/978-3-642-25274-7\_14",

language = "English",

isbn = "9783642252730",

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

publisher = "Springer Verlag",

pages = "134--144",

booktitle = "Advances in Artificial Intelligence - 14th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2011, Proceedings",

address = "Germany",

note = "14th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2011 ; Conference date: 07-11-2011 Through 11-11-2011",

}

Hernández-González, J & Inza, I 2011, Learning naive Bayes models for multiple-instance learning with label proportions. in Advances in Artificial Intelligence - 14th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2011, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 7023 LNAI, Springer Verlag, pp. 134-144, 14th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2011, La Laguna, Spain, 7/11/11. https://doi.org/10.1007/978-3-642-25274-7_14

Learning naive Bayes models for multiple-instance learning with label proportions. / Hernández-González, Jerónimo; Inza, Iñaki.
Advances in Artificial Intelligence - 14th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2011, Proceedings. Springer Verlag, 2011. p. 134-144 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7023 LNAI).

Research output: Chapter in Book/Conference proceeding › Conference proceeding › peer-review

TY - GEN

T1 - Learning naive Bayes models for multiple-instance learning with label proportions

AU - Hernández-González, Jerónimo

AU - Inza, Iñaki

PY - 2011

Y1 - 2011

N2 - This paper deals with the problem of multi-instance learning when label proportions are provided. In this classification problem, the instances of the dataset are divided into disjoint groups, where there is no certainty about the labels associated with individual samples. However, in each group the number of instances that belong to each class is known. We propose several versions of an EM-algorithm that learns naive Bayes models to deal with the exposed problem. The proposed algorithms are evaluated on synthetic and real datasets, and compared with state-of-the-art approaches. The obtained results show a competitive behaviour of our proposals.

AB - This paper deals with the problem of multi-instance learning when label proportions are provided. In this classification problem, the instances of the dataset are divided into disjoint groups, where there is no certainty about the labels associated with individual samples. However, in each group the number of instances that belong to each class is known. We propose several versions of an EM-algorithm that learns naive Bayes models to deal with the exposed problem. The proposed algorithms are evaluated on synthetic and real datasets, and compared with state-of-the-art approaches. The obtained results show a competitive behaviour of our proposals.

KW - EM algorithm

KW - Multiple-instance learning with label proportions

KW - Naive Bayes

KW - supervised classification

UR - https://www.scopus.com/pages/publications/81055149948

U2 - 10.1007/978-3-642-25274-7_14

DO - 10.1007/978-3-642-25274-7_14

M3 - Conference proceeding

AN - SCOPUS:81055149948

SN - 9783642252730

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

SP - 134

EP - 144

BT - Advances in Artificial Intelligence - 14th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2011, Proceedings

PB - Springer Verlag

T2 - 14th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2011

Y2 - 7 November 2011 through 11 November 2011

ER -

Hernández-González J, Inza I. Learning naive Bayes models for multiple-instance learning with label proportions. In Advances in Artificial Intelligence - 14th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2011, Proceedings. Springer Verlag. 2011. p. 134-144. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-642-25274-7_14

Learning naive Bayes models for multiple-instance learning with label proportions

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this