Hierarchical fuzzy rule based systems using an information theoretic approach

Waldock, Antony; Carse, Brian; Melhuish, Chris

Abstract

This paper proposes a novel anytime algorithm for the construction of a Hierarchical Fuzzy Rule Based System using an information theoretic approach to specialise rules that do not effectively model the decision space. The amount of uncertainty tolerated within the decision provides a single tuneable parameter to control the trade off between accuracy and interpretability. The algorithm is empirically compared with existing methods of function approximation and is demonstrated on a mobile robot application in simulation. © Springer-Verlag 2006.

• Journal Article Type: Article
• Publication Date: Aug 1, 2006
• Journal: Soft Computing
• Print ISSN: 1432-7643
• Electronic ISSN: 1433-7479
• Publisher: Springer (part of Springer Nature)
• Peer Reviewed: Not Peer Reviewed
• Volume: 10
• Issue: 10
• Pages: 867-879
• DOI: https://doi.org/10.1007/s00500-005-0013-y
• Keywords: information theory, hierarchical fuzzy rule based systems, mobile robot
• Public URL: https://uwe-repository.worktribe.com/output/1040595
• Publisher URL: http://dx.doi.org/10.1007/s00500-005-0013-y
• Additional Information: Additional Information : The accuracy/interpretability trade-off in fuzzy systems development is a well known problem. In particular, interpretability is a serious requirement if fuzzy rule-based systems are to be adopted for industrial applications. The work describes and evaluates a new approach based on information theory where the amount of uncertainty tolerated within fuzzy decision-making decision provides a single tuneable parameter to control the trade off between accuracy and interpretability.