Evolving possibilistic fuzzy modeling

Abstract

This paper suggests an evolving possibilistic approach for fuzzy modeling of time-varying processes. The approach is based on an extension of the possibilistic fuzzy c-means clustering and functional fuzzy rule-based modeling. Evolving possibilistic fuzzy modeling (ePFM) employs memberships and typicalities to recursively cluster data, and uses participatory learning to adapt the model structure as a stream data is input. Data produced by a synthetic time-varying process with parameter drift is used to show the usefulness, and to highlight the performance of the ePFM when compared with state of the art evolving fuzzy, neural, and neural-fuzzy modeling approaches. The results show that ePFM is a potential candidate for nonlinear time varying systems modeling, with comparable or better performance than alternative approaches, mainly when noise and outliers affect the data available.