New approach based on fuzzy hypergraphs in granular computing (an application to the urban vulnerability assessment)

Classifying objects based on the simultaneous impact of various parameters has always been challenging due to heterogeneity, impact conflict, and sometimes parameter uncertainty. The purpose of this study is to provide a method for classifying such data. In the proposed method, fuzzy hypergraphs were used to define the granular structures in order to apply the simultaneous effect of heterogeneous and weighted parameters in the classification. This method has been implemented and validated on Fisher’s intuitive research in relation to the classification of iris flowers. Evaluation and comparison of the proposed method with Fisher’s experimental results showed higher efficiency and accuracy in flower classification. The proposed method has been used to assess the seismic risk of 50,000 buildings based on 10 heterogeneous parameters. Seismic risk classification showed that more than 88% of buildings were classified, and 12% of buildings that could not be classified due to excessive scatter of parameter values were classified using a very small confidence radius. The results indicate the ability of the proposed method to classify objects with the least similarity and number of effective parameters in classification.