Fuzzy C-Means and Two Clusters' Centers Method for Generating Interval Type-2 Membership Function

Existing works investigated the construction of fuzzy type-1 (FT1) membership function (MF). However, recent findings show that interval type-2 (IT2)-based fuzzy inference system (FIS) is found to be more accurate and precise than FT1. Hence, the research on how to generate IT2 MF from data is significant to be conducted. Besides, existing works also investigated the construction of IT2 MF using IT2 Fuzzy C-Means (FCM) method. The evident shows that the construction of IT2 MF from IT2 FCM method may not suitable for all kind of data sets. Hence, the objectives of this paper are to present a methodology for the generation of IT2 MF using general FCM (non-IT2 FCM) data clustering method and to describe the implementation of the proposed IT2 MFs in an FIS. The experiment results show that IT2 MFs have successfully been constructed by using general FCM and two clusters' centers approach.