BNC Structure Learning: G2 Algorithm based Heuristic

Author

CHENG, Zekai ; Qin, Feng ; Yang, Bo

Author_Institution

Sch. of Comput. Sci., Anhui Univ. of Technol., Maanshan

Volume

2

fYear

0

fDate

0-0 0

Firstpage

6010

Lastpage

6014

Abstract

Structure learning for Bayesian networks classifier is NP-hard problem, K2 algorithm is one of efficacious and accurate algorithm. K2 algorithm confirms the order of nodes firstly. To a certain extent this limits in non-information. This paper purposes a new heuristic Bayesian networks classifier structure learning G2 algorithm. G2 algorithm use NB and TAN structure which learns as heuristic information, using K2 algorithm learning BNC structure. The experimental result shows that G2 algorithm can solve nodes order in non-information. Arcs are sententious, compare NB and TAN structure, it´s more reasonable, classification accuracy improves, this algorithm adapts many aspects

Keywords

belief networks; computational complexity; learning (artificial intelligence); pattern classification; BNC structure learning; Bayesian networks classifier; G2 algorithm based heuristic; K2 algorithm; NB structure; NP-hard problem; TAN structure; heuristic search; Automation; Bayesian methods; Computer networks; Electronic mail; Heuristic algorithms; Intelligent control; NP-hard problem; Niobium compounds; BNC; Bayesian Networks; Heuristic Search; Structure Learning;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Control and Automation, 2006. WCICA 2006. The Sixth World Congress on

Conference_Location

Dalian

Print_ISBN

1-4244-0332-4

Type

conf

DOI

10.1109/WCICA.2006.1714233

Filename

1714233