Transparent linguistic interface generation and its application in fuzzy decision trees

The linguistic interface is a group of fuzzy sets and their corresponding membership functions applied to define the linguistic terms used in fuzzy modeling, granular computing and computing with words. The fuzzy C-means algorithm (FCM) is widely used in the generation of membership functions for linguistic terms from historical data. However, most of FCM-based membership function generation algorithms and their variants consider little on the transparency or the understandability of the resulting membership functions. This paper proposes a genetic pre-shaped fuzzy C-means algorithm (GPFCM) to generate transparent membership functions for linguistic terms in linguistic interfaces. The proposed algorithm will preserve predefined transparent shapes of membership functions during the optimization process of the clustering algorithm. To avoid local optimality, the genetic algorithm is applied to solve the optimization problem in the clustering algorithm. Numeric experiments based on data collected in a real civil project demonstrate the feasibility and superiority of the proposed new algorithm. The classification problems solved by fuzzy decision trees based on different linguistic interfaces are provided to demonstrate the advantages of the proposed linguistic interface generation method.