Mining and visualising ordinal data with non-parametric continuous BBNs

AM Hanea, Dorota Kurowicka, Roger M Cooke, DA Ababei

Research output: Contribution to journalArticle

32 Citations (Scopus)

Abstract

Data mining is the process of extracting and analysing information from large databases. Graphical models are a suitable framework for probabilistic modelling. A Bayesian Belief Net (BBN) is a probabilistic graphical model, which represents joint distributions in an intuitive and efficient way. It encodes the probability density (or mass) function of a set of variables by specifying a number of conditional independence statements in the form of a directed acyclic graph. Specifying the structure of the model is one of the most important design choices in graphical modelling. Notwithstanding their potential, there are only a limited number of applications of graphical models on very complex and large databases. A method for mining ordinal multivariate data using non-parametric BBNs is presented. The main advantage of this method is that it can handle a large number of continuous variables, without making any assumptions about their marginal distributions, in a very fast manner. Once the BBN is learned from data, it can be further used for prediction. This approach allows for rapid conditionalisation, which is a very important feature of a BBN from a user's standpoint. © 2008 Elsevier B.V. All rights reserved.
LanguageEnglish
Pages668-687
Number of pages20
JournalComputational Statistics and Data Analysis
Volume54
Issue number3
DOIs
Publication statusPublished - 1 Mar 2010

Fingerprint

Ordinal Data
Graphical Models
Mining
Graphical Modeling
Probabilistic Modeling
Conditional Independence
Directed Acyclic Graph
Multivariate Data
Continuous Variables
Marginal Distribution
Probability Density
Joint Distribution
Probabilistic Model
Intuitive
Data Mining
Data mining
Prediction
Beliefs
Graphical models
Ordinal data

Keywords

  • conditional independence statements
  • continuous variables
  • graphical modelling
  • probabilistic modelling
  • probability density function
  • Bayesian networks

Cite this

Hanea, AM ; Kurowicka, Dorota ; Cooke, Roger M ; Ababei, DA. / Mining and visualising ordinal data with non-parametric continuous BBNs. In: Computational Statistics and Data Analysis. 2010 ; Vol. 54, No. 3. pp. 668-687.
@article{5336f8d3a0a445cbbf21f49d66d7842f,
title = "Mining and visualising ordinal data with non-parametric continuous BBNs",
abstract = "Data mining is the process of extracting and analysing information from large databases. Graphical models are a suitable framework for probabilistic modelling. A Bayesian Belief Net (BBN) is a probabilistic graphical model, which represents joint distributions in an intuitive and efficient way. It encodes the probability density (or mass) function of a set of variables by specifying a number of conditional independence statements in the form of a directed acyclic graph. Specifying the structure of the model is one of the most important design choices in graphical modelling. Notwithstanding their potential, there are only a limited number of applications of graphical models on very complex and large databases. A method for mining ordinal multivariate data using non-parametric BBNs is presented. The main advantage of this method is that it can handle a large number of continuous variables, without making any assumptions about their marginal distributions, in a very fast manner. Once the BBN is learned from data, it can be further used for prediction. This approach allows for rapid conditionalisation, which is a very important feature of a BBN from a user's standpoint. {\circledC} 2008 Elsevier B.V. All rights reserved.",
keywords = "conditional independence statements, continuous variables , graphical modelling, probabilistic modelling, probability density function, Bayesian networks",
author = "AM Hanea and Dorota Kurowicka and Cooke, {Roger M} and DA Ababei",
year = "2010",
month = "3",
day = "1",
doi = "10.1016/j.csda.2008.09.032",
language = "English",
volume = "54",
pages = "668--687",
journal = "Computational Statistics and Data Analysis",
issn = "0167-9473",
number = "3",

}

Mining and visualising ordinal data with non-parametric continuous BBNs. / Hanea, AM; Kurowicka, Dorota; Cooke, Roger M; Ababei, DA.

In: Computational Statistics and Data Analysis, Vol. 54, No. 3, 01.03.2010, p. 668-687.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Mining and visualising ordinal data with non-parametric continuous BBNs

AU - Hanea, AM

AU - Kurowicka, Dorota

AU - Cooke, Roger M

AU - Ababei, DA

PY - 2010/3/1

Y1 - 2010/3/1

N2 - Data mining is the process of extracting and analysing information from large databases. Graphical models are a suitable framework for probabilistic modelling. A Bayesian Belief Net (BBN) is a probabilistic graphical model, which represents joint distributions in an intuitive and efficient way. It encodes the probability density (or mass) function of a set of variables by specifying a number of conditional independence statements in the form of a directed acyclic graph. Specifying the structure of the model is one of the most important design choices in graphical modelling. Notwithstanding their potential, there are only a limited number of applications of graphical models on very complex and large databases. A method for mining ordinal multivariate data using non-parametric BBNs is presented. The main advantage of this method is that it can handle a large number of continuous variables, without making any assumptions about their marginal distributions, in a very fast manner. Once the BBN is learned from data, it can be further used for prediction. This approach allows for rapid conditionalisation, which is a very important feature of a BBN from a user's standpoint. © 2008 Elsevier B.V. All rights reserved.

AB - Data mining is the process of extracting and analysing information from large databases. Graphical models are a suitable framework for probabilistic modelling. A Bayesian Belief Net (BBN) is a probabilistic graphical model, which represents joint distributions in an intuitive and efficient way. It encodes the probability density (or mass) function of a set of variables by specifying a number of conditional independence statements in the form of a directed acyclic graph. Specifying the structure of the model is one of the most important design choices in graphical modelling. Notwithstanding their potential, there are only a limited number of applications of graphical models on very complex and large databases. A method for mining ordinal multivariate data using non-parametric BBNs is presented. The main advantage of this method is that it can handle a large number of continuous variables, without making any assumptions about their marginal distributions, in a very fast manner. Once the BBN is learned from data, it can be further used for prediction. This approach allows for rapid conditionalisation, which is a very important feature of a BBN from a user's standpoint. © 2008 Elsevier B.V. All rights reserved.

KW - conditional independence statements

KW - continuous variables

KW - graphical modelling

KW - probabilistic modelling

KW - probability density function

KW - Bayesian networks

U2 - 10.1016/j.csda.2008.09.032

DO - 10.1016/j.csda.2008.09.032

M3 - Article

VL - 54

SP - 668

EP - 687

JO - Computational Statistics and Data Analysis

T2 - Computational Statistics and Data Analysis

JF - Computational Statistics and Data Analysis

SN - 0167-9473

IS - 3

ER -