Approximate Reasoning with Fuzzy Rule-based Systems Supported by Transformation-based Interpolative Reasoning

Chen, T. (Keynote speaker)

Activity: Talk or presentation typesInvited talk

Description

As a significant area of computation intelligence, the linguistically inspired fuzzy rule-based systems are widely known for the approximate inference mechanism mimicking human reasoning and the strong capability in dealing with imprecision and real-world uncertainty, making it an ideal tool to design interpretable and intelligent models by allowing non-technical domain experts to understand and interrogate the underlying systems employed. Furthermore, in the presence of insufficient knowledge or sparse rule bases, where an input observation may not match and fire any of the existing fuzzy rules, thereby leading to no conclusion, fuzzy interpolative reasoning plays a significant role in resolving this restriction for fuzzy systems. This talk will start with the motivation and overview of a fuzzy system for decision making, before introducing recent advances in fuzzy system development. This is then followed by the introduction of fuzzy interpolative reasoning in supporting fuzzy system design and of its advances in transformation-based fuzzy rule interpolation. The talk will conclude with an outline of opportunities for future development.
Period26 Dec 2021
Event title3rd IEEE International Conference on Power Data Science 2021
Event typeConference
Conference number3
Degree of RecognitionInternational