Learning Bayesian network structures by searching for the best ordering with genetic algorithms

Author

Larranaga, Pedro ; Kuijpers, Cindy M H ; Murga, Roberto H. ; Yurramendi, Yosu

Author_Institution

Dept. of Comput. Sci. & Artificial Intelligence, Univ. of the Basque Country, San Sebastian, Spain

Volume

26

Issue

4

fYear

1996

fDate

7/1/1996 12:00:00 AM

Firstpage

487

Lastpage

493

Abstract

Presents a new methodology for inducing Bayesian network structures from a database of cases. The methodology is based on searching for the best ordering of the system variables by means of genetic algorithms. Since this problem of finding an optimal ordering of variables resembles the traveling salesman problem, the authors use genetic operators that were developed for the latter problem. The quality of a variable ordering is evaluated with the structure-learning algorithm K2. The authors present empirical results that were obtained with a simulation of the ALARM network

Keywords

directed graphs; genetic algorithms; inference mechanisms; learning (artificial intelligence); search problems; ALARM network; Bayesian network structures; K2 structure-learning algorithm; best ordering; genetic algorithms; optimal ordering; searching; traveling salesman problem; Artificial intelligence; Bayesian methods; Computer science; Databases; Genetic algorithms; Graphical models; Humans; Probability distribution; Random variables; Traveling salesman problems;

fLanguage

English

Journal_Title

Systems, Man and Cybernetics, Part A: Systems and Humans, IEEE Transactions on

Publisher

ieee

ISSN

1083-4427

Type

jour

DOI

10.1109/3468.508827

Filename

508827