A Design of CGK-Based Granular Model Using Hierarchical Structure

In this paper, we propose context-based GK clustering and design a CGK-based granular model hierarchical model. Existing fuzzy generates clusters using Euclidean distances. However, there is problem in that performance decreases when cluster created from data with strong nonlinearity. To improve problem, used. creates Mahalanobis distance. (CGK) clustering, which adds method considers the output space existing to create not only input but also space. is. Based on proposed CGK were designed. Since of form context, it has advantage verbally expressing prediction result, structure can generate high-dimensional information granules, so meaningful high abstraction value granules be created. order verify validity as result conducting an experiment concrete compressive strength database, was confirmed methods showed superior than models.