Ensemble bayesian networks evolved with speciation for high-performance prediction in data mining

Kyung Joong Kim, Sung-Bae Cho

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Bayesian networks (BNs) can be easily refined (or learn) using data given prior knowledge about a changing environment. Furthermore, by exploring multiple diverse BNs in parallel, it is expected that an intelligent system may adapt quickly to changes in the environment, resulting in robust prediction. Recently, there have been attempts to design BN structures using evolutionary algorithms; however, most of these have used only the fittest solution from the final generation. Because it is difficult to combine all of the important factors into a single evaluation function, the solution is often biased and of limited adaptability. Here we describe a method of generating diverse BN structures via speciation and selective combination for adaptive prediction. Experiments using the seven benchmark networks show that the proposed method can result in improved accuracy in handling uncertainty by exploiting ensembles of BNs evolved by speciation.

Original languageEnglish
Pages (from-to)1065-1080
Number of pages16
JournalSoft Computing
Volume21
Issue number4
DOIs
Publication statusPublished - 2017 Feb 1

Fingerprint

Speciation
Performance Prediction
Bayesian networks
Bayesian Networks
Data mining
Data Mining
Ensemble
High Performance
Network Structure
Function evaluation
Prediction
Intelligent systems
Evaluation Function
Intelligent Systems
Adaptability
Prior Knowledge
Evolutionary algorithms
Biased
Evolutionary Algorithms
Benchmark

All Science Journal Classification (ASJC) codes

  • Software
  • Theoretical Computer Science
  • Geometry and Topology

Cite this

@article{50b9c596fcd04b7794624a27bcdc1292,
title = "Ensemble bayesian networks evolved with speciation for high-performance prediction in data mining",
abstract = "Bayesian networks (BNs) can be easily refined (or learn) using data given prior knowledge about a changing environment. Furthermore, by exploring multiple diverse BNs in parallel, it is expected that an intelligent system may adapt quickly to changes in the environment, resulting in robust prediction. Recently, there have been attempts to design BN structures using evolutionary algorithms; however, most of these have used only the fittest solution from the final generation. Because it is difficult to combine all of the important factors into a single evaluation function, the solution is often biased and of limited adaptability. Here we describe a method of generating diverse BN structures via speciation and selective combination for adaptive prediction. Experiments using the seven benchmark networks show that the proposed method can result in improved accuracy in handling uncertainty by exploiting ensembles of BNs evolved by speciation.",
author = "Kim, {Kyung Joong} and Sung-Bae Cho",
year = "2017",
month = "2",
day = "1",
doi = "10.1007/s00500-015-1841-z",
language = "English",
volume = "21",
pages = "1065--1080",
journal = "Soft Computing",
issn = "1432-7643",
publisher = "Springer Verlag",
number = "4",

}

Ensemble bayesian networks evolved with speciation for high-performance prediction in data mining. / Kim, Kyung Joong; Cho, Sung-Bae.

In: Soft Computing, Vol. 21, No. 4, 01.02.2017, p. 1065-1080.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Ensemble bayesian networks evolved with speciation for high-performance prediction in data mining

AU - Kim, Kyung Joong

AU - Cho, Sung-Bae

PY - 2017/2/1

Y1 - 2017/2/1

N2 - Bayesian networks (BNs) can be easily refined (or learn) using data given prior knowledge about a changing environment. Furthermore, by exploring multiple diverse BNs in parallel, it is expected that an intelligent system may adapt quickly to changes in the environment, resulting in robust prediction. Recently, there have been attempts to design BN structures using evolutionary algorithms; however, most of these have used only the fittest solution from the final generation. Because it is difficult to combine all of the important factors into a single evaluation function, the solution is often biased and of limited adaptability. Here we describe a method of generating diverse BN structures via speciation and selective combination for adaptive prediction. Experiments using the seven benchmark networks show that the proposed method can result in improved accuracy in handling uncertainty by exploiting ensembles of BNs evolved by speciation.

AB - Bayesian networks (BNs) can be easily refined (or learn) using data given prior knowledge about a changing environment. Furthermore, by exploring multiple diverse BNs in parallel, it is expected that an intelligent system may adapt quickly to changes in the environment, resulting in robust prediction. Recently, there have been attempts to design BN structures using evolutionary algorithms; however, most of these have used only the fittest solution from the final generation. Because it is difficult to combine all of the important factors into a single evaluation function, the solution is often biased and of limited adaptability. Here we describe a method of generating diverse BN structures via speciation and selective combination for adaptive prediction. Experiments using the seven benchmark networks show that the proposed method can result in improved accuracy in handling uncertainty by exploiting ensembles of BNs evolved by speciation.

UR - http://www.scopus.com/inward/record.url?scp=84939626031&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84939626031&partnerID=8YFLogxK

U2 - 10.1007/s00500-015-1841-z

DO - 10.1007/s00500-015-1841-z

M3 - Article

AN - SCOPUS:84939626031

VL - 21

SP - 1065

EP - 1080

JO - Soft Computing

JF - Soft Computing

SN - 1432-7643

IS - 4

ER -