A dynamic Bayesian Network Model for Hierarchical Classification and its Application in Predicting Yeast Genes Functions

Xutao Deng, Huimin Geng, Hesham H Ali

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

Abstract

In this paper, we propose a Dynamic Naive Bayesian (DNB) network model for classifying data sets with hierarchical labels. The DNB model is built upon a Naive Bayesian (NB) network, a successful classifier for data with flattened (nonhierarchical) class labels. The problems using flattened class labels for hierarchical classification are addressed in this paper. The DNB has a top-down structure with each level of the class hierarchy modeled as a random variable. We defined augmenting operations to transform class hierarchy into a form that satisfies the probability law. We present algorithms for efficient learning and inference with the DNB model. The learning algorithm can be used to estimate the parameters of the network. The inference algorithm is designed to find the optimal classification path in the class hierarchy. The methods are tested on yeast gene expression data sets, and the classification accuracy with DNB classifier is significantly higher than it is with previous approaches- flattened classification using NB classifier.

Original languageEnglish (US)
Title of host publicationAssociation for Information Systems - 11th Americas Conference on Information Systems, AMCIS 2005
Subtitle of host publicationA Conference on a Human Scale
Pages2626-2634
Number of pages9
StatePublished - Dec 1 2005
Event11th Americas Conference on Information Systems, AMCIS 2005 - Omaha, NE, United States
Duration: Aug 11 2005Aug 15 2005

Publication series

NameAssociation for Information Systems - 11th Americas Conference on Information Systems, AMCIS 2005: A Conference on a Human Scale
Volume6

Conference

Conference11th Americas Conference on Information Systems, AMCIS 2005
CountryUnited States
CityOmaha, NE
Period8/11/058/15/05

Fingerprint

Bayesian networks
Yeast
Genes
Labels
Classifiers
learning
Random variables
Gene expression
Learning algorithms
Law

Keywords

  • Bayesian network
  • Dynamic Bayesian network
  • Hierarchical classification
  • Naive Bayesian classifier

ASJC Scopus subject areas

  • Computer Science Applications
  • Computer Networks and Communications
  • Information Systems
  • Library and Information Sciences

Cite this

Deng, X., Geng, H., & Ali, H. H. (2005). A dynamic Bayesian Network Model for Hierarchical Classification and its Application in Predicting Yeast Genes Functions. In Association for Information Systems - 11th Americas Conference on Information Systems, AMCIS 2005: A Conference on a Human Scale (pp. 2626-2634). (Association for Information Systems - 11th Americas Conference on Information Systems, AMCIS 2005: A Conference on a Human Scale; Vol. 6).

A dynamic Bayesian Network Model for Hierarchical Classification and its Application in Predicting Yeast Genes Functions. / Deng, Xutao; Geng, Huimin; Ali, Hesham H.

Association for Information Systems - 11th Americas Conference on Information Systems, AMCIS 2005: A Conference on a Human Scale. 2005. p. 2626-2634 (Association for Information Systems - 11th Americas Conference on Information Systems, AMCIS 2005: A Conference on a Human Scale; Vol. 6).

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

Deng, X, Geng, H & Ali, HH 2005, A dynamic Bayesian Network Model for Hierarchical Classification and its Application in Predicting Yeast Genes Functions. in Association for Information Systems - 11th Americas Conference on Information Systems, AMCIS 2005: A Conference on a Human Scale. Association for Information Systems - 11th Americas Conference on Information Systems, AMCIS 2005: A Conference on a Human Scale, vol. 6, pp. 2626-2634, 11th Americas Conference on Information Systems, AMCIS 2005, Omaha, NE, United States, 8/11/05.
Deng X, Geng H, Ali HH. A dynamic Bayesian Network Model for Hierarchical Classification and its Application in Predicting Yeast Genes Functions. In Association for Information Systems - 11th Americas Conference on Information Systems, AMCIS 2005: A Conference on a Human Scale. 2005. p. 2626-2634. (Association for Information Systems - 11th Americas Conference on Information Systems, AMCIS 2005: A Conference on a Human Scale).
Deng, Xutao ; Geng, Huimin ; Ali, Hesham H. / A dynamic Bayesian Network Model for Hierarchical Classification and its Application in Predicting Yeast Genes Functions. Association for Information Systems - 11th Americas Conference on Information Systems, AMCIS 2005: A Conference on a Human Scale. 2005. pp. 2626-2634 (Association for Information Systems - 11th Americas Conference on Information Systems, AMCIS 2005: A Conference on a Human Scale).
@inproceedings{0ac57654734c405496cbdace05d4f54b,
title = "A dynamic Bayesian Network Model for Hierarchical Classification and its Application in Predicting Yeast Genes Functions",
abstract = "In this paper, we propose a Dynamic Naive Bayesian (DNB) network model for classifying data sets with hierarchical labels. The DNB model is built upon a Naive Bayesian (NB) network, a successful classifier for data with flattened (nonhierarchical) class labels. The problems using flattened class labels for hierarchical classification are addressed in this paper. The DNB has a top-down structure with each level of the class hierarchy modeled as a random variable. We defined augmenting operations to transform class hierarchy into a form that satisfies the probability law. We present algorithms for efficient learning and inference with the DNB model. The learning algorithm can be used to estimate the parameters of the network. The inference algorithm is designed to find the optimal classification path in the class hierarchy. The methods are tested on yeast gene expression data sets, and the classification accuracy with DNB classifier is significantly higher than it is with previous approaches- flattened classification using NB classifier.",
keywords = "Bayesian network, Dynamic Bayesian network, Hierarchical classification, Naive Bayesian classifier",
author = "Xutao Deng and Huimin Geng and Ali, {Hesham H}",
year = "2005",
month = "12",
day = "1",
language = "English (US)",
isbn = "9781604235531",
series = "Association for Information Systems - 11th Americas Conference on Information Systems, AMCIS 2005: A Conference on a Human Scale",
pages = "2626--2634",
booktitle = "Association for Information Systems - 11th Americas Conference on Information Systems, AMCIS 2005",

}

TY - GEN

T1 - A dynamic Bayesian Network Model for Hierarchical Classification and its Application in Predicting Yeast Genes Functions

AU - Deng, Xutao

AU - Geng, Huimin

AU - Ali, Hesham H

PY - 2005/12/1

Y1 - 2005/12/1

N2 - In this paper, we propose a Dynamic Naive Bayesian (DNB) network model for classifying data sets with hierarchical labels. The DNB model is built upon a Naive Bayesian (NB) network, a successful classifier for data with flattened (nonhierarchical) class labels. The problems using flattened class labels for hierarchical classification are addressed in this paper. The DNB has a top-down structure with each level of the class hierarchy modeled as a random variable. We defined augmenting operations to transform class hierarchy into a form that satisfies the probability law. We present algorithms for efficient learning and inference with the DNB model. The learning algorithm can be used to estimate the parameters of the network. The inference algorithm is designed to find the optimal classification path in the class hierarchy. The methods are tested on yeast gene expression data sets, and the classification accuracy with DNB classifier is significantly higher than it is with previous approaches- flattened classification using NB classifier.

AB - In this paper, we propose a Dynamic Naive Bayesian (DNB) network model for classifying data sets with hierarchical labels. The DNB model is built upon a Naive Bayesian (NB) network, a successful classifier for data with flattened (nonhierarchical) class labels. The problems using flattened class labels for hierarchical classification are addressed in this paper. The DNB has a top-down structure with each level of the class hierarchy modeled as a random variable. We defined augmenting operations to transform class hierarchy into a form that satisfies the probability law. We present algorithms for efficient learning and inference with the DNB model. The learning algorithm can be used to estimate the parameters of the network. The inference algorithm is designed to find the optimal classification path in the class hierarchy. The methods are tested on yeast gene expression data sets, and the classification accuracy with DNB classifier is significantly higher than it is with previous approaches- flattened classification using NB classifier.

KW - Bayesian network

KW - Dynamic Bayesian network

KW - Hierarchical classification

KW - Naive Bayesian classifier

UR - http://www.scopus.com/inward/record.url?scp=84870040095&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84870040095&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84870040095

SN - 9781604235531

T3 - Association for Information Systems - 11th Americas Conference on Information Systems, AMCIS 2005: A Conference on a Human Scale

SP - 2626

EP - 2634

BT - Association for Information Systems - 11th Americas Conference on Information Systems, AMCIS 2005

ER -