Optimization Models for Training Belief-Rule-Based Systems

J Liu, J Wang, D-L Xu, J-B Yang

Research output: Contribution to journalArticlepeer-review

226 Citations (Scopus)
Original languageEnglish
Pages (from-to)569-585
JournalIEEE Transactions on Systems, Man, and Cybernetics, Part A: Systems and Humans
Issue number4
Publication statusPublished (in print/issue) - 1 Jul 2007

Bibliographical note

Other Details
It is difficult to accurately determine rule-bases entirely subjectively, particularly for a large-scale rule-base. This paper presents optimisation models for a newly developed belief rule-base system. The models provide a non-black-box simulator for which input/output information can be incomplete or vague, either numerical or judgmental, or mixed. Application to analyse safety of engineering systems and leak detection in oil pipelines has shown their abilities to simulate real situations in a meaningful, efficient, consistent, and optimal way. The work relates to developments in two projects funded by EPSRC (GR/S85498/01) and DEFRA (AFM222) respectively, held in the University of Manchester.

Cite this