Title :
A Multi-heuristic Cooperative Ant Colony System for Optimizing Elimination Ordering of Bayesian Networks
Author :
Dong, Xuchu ; Ouyang, Dantong ; Ye, Yuxin ; Yu, Haihong ; Zhang, Yonggang
Author_Institution :
Dept. of Comput. Sci. & Technol., Jilin Univ., Jilin, China
Abstract :
To solve the problem of searching for an optimal elimination ordering of Bayesian networks, a novel effective heuristic, MinSum Weight, and an ACS approach incorporated with multi-heuristic mechanism are proposed. The ACS approach named MHC-ACS utilizes a set of heuristics to direct the ants moving in the search space. The cooperation of multiple heuristics helps ants explore more regions. Moreover, the most appropriate heuristic will be identified and be reinforced with the evolution of the whole system. Experiments demonstrate that MHC-ACS has a better performance than other swarm intelligence methods.
Keywords :
belief networks; optimisation; Bayesian networks; MinSum weight; multiheuristic cooperative ant colony system; multiheuristic mechanism; optimizing elimination ordering; swarm intelligence methods; Bayesian network; ant colony system; elimination ordering; heuristics; junction tree;
Conference_Titel :
Web Intelligence and Intelligent Agent Technology (WI-IAT), 2010 IEEE/WIC/ACM International Conference on
Print_ISBN :
978-1-4244-8482-9
Electronic_ISBN :
978-0-7695-4191-4
DOI :
10.1109/WI-IAT.2010.33
