Discover knowledge from distribution maps using bayesian networks

Norazwin Buang; Nianjun Liu; Terry Caelli; Rob Lesslie; Michael J. Hill

Discover knowledge from distribution maps using bayesian networks

Norazwin Buang^*, Nianjun Liu, Terry Caelli, Rob Lesslie, Michael J. Hill

^*Corresponding author for this work

Research output: Contribution to journal › Conference article › peer-review

1 Citation (Scopus)

Abstract

This paper applies a Bayesian network to model multi criteria distribution maps and to discover knowledge contained in spatial data. The procedure consists of three steps: pre processing map data, training the Bayesian Network model using distribution maps of Australia and testing the generalization and diagnosis of the model using individual states' maps. The Bayesian network that we used in this study is known as naïve Bayesian network. Results show that this environmental Bayesian network model can generalize the classification rules from training data for good prediction and diagnosis of a distribution map.

Original language	English
Pages (from-to)	69-74
Number of pages	6
Journal	Conferences in Research and Practice in Information Technology Series
Volume	61
Publication status	Published - 2006
Event	5th Australasian Data Mining Conference, AusDM 2006 - Sydney, NSW, Australia Duration: 29 Nov 2006 → 30 Nov 2006

Cite this

@article{60e6954d0d6a43d3b0071f85f4b12ecd,

title = "Discover knowledge from distribution maps using bayesian networks",

abstract = "This paper applies a Bayesian network to model multi criteria distribution maps and to discover knowledge contained in spatial data. The procedure consists of three steps: pre processing map data, training the Bayesian Network model using distribution maps of Australia and testing the generalization and diagnosis of the model using individual states' maps. The Bayesian network that we used in this study is known as na{\"i}ve Bayesian network. Results show that this environmental Bayesian network model can generalize the classification rules from training data for good prediction and diagnosis of a distribution map.",

keywords = "Bayesian network, Combining evidence, Distribution maps, Multi-criteria analysis",

author = "Norazwin Buang and Nianjun Liu and Terry Caelli and Rob Lesslie and Hill, {Michael J.}",

year = "2006",

language = "English",

volume = "61",

pages = "69--74",

journal = "Conferences in Research and Practice in Information Technology Series",

issn = "1445-1336",

note = "5th Australasian Data Mining Conference, AusDM 2006 ; Conference date: 29-11-2006 Through 30-11-2006",

}

TY - JOUR

T1 - Discover knowledge from distribution maps using bayesian networks

AU - Buang, Norazwin

AU - Liu, Nianjun

AU - Caelli, Terry

AU - Lesslie, Rob

AU - Hill, Michael J.

PY - 2006

Y1 - 2006

N2 - This paper applies a Bayesian network to model multi criteria distribution maps and to discover knowledge contained in spatial data. The procedure consists of three steps: pre processing map data, training the Bayesian Network model using distribution maps of Australia and testing the generalization and diagnosis of the model using individual states' maps. The Bayesian network that we used in this study is known as naïve Bayesian network. Results show that this environmental Bayesian network model can generalize the classification rules from training data for good prediction and diagnosis of a distribution map.

AB - This paper applies a Bayesian network to model multi criteria distribution maps and to discover knowledge contained in spatial data. The procedure consists of three steps: pre processing map data, training the Bayesian Network model using distribution maps of Australia and testing the generalization and diagnosis of the model using individual states' maps. The Bayesian network that we used in this study is known as naïve Bayesian network. Results show that this environmental Bayesian network model can generalize the classification rules from training data for good prediction and diagnosis of a distribution map.

KW - Bayesian network

KW - Combining evidence

KW - Distribution maps

KW - Multi-criteria analysis

UR - http://www.scopus.com/inward/record.url?scp=84870488164&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:84870488164

SN - 1445-1336

VL - 61

SP - 69

EP - 74

JO - Conferences in Research and Practice in Information Technology Series

JF - Conferences in Research and Practice in Information Technology Series

T2 - 5th Australasian Data Mining Conference, AusDM 2006

Y2 - 29 November 2006 through 30 November 2006

ER -

Discover knowledge from distribution maps using bayesian networks

Abstract

Other files and links

Fingerprint

Cite this