Identification of Strongly Correlated n-Tuples with the Use of Fuzzy Equivalence Relations
摘要
In several of the real-world applications, it is common, that relationship between tuples of attributes of dimension higher than two needs to be examined. Since there is need to examine relationships in multidimensional datasets analysts use various approaches to do so − one class of such methods, labelled as correlation methods, focuses on analysis of the strength and direction of functional relationships between attributes in an attribute pair. The main objective of this paper is motivated by possibility of inclusion of fuzzy equivalence relations to correlation matrices for the purposes of identification of n-tuples of attributes with significant relationships where \(n>2\) . For the evaluation purposes of the proposed fuzzy equivalence relation-based method, the dataset focused on renewable energy is used and the obtained results are compared with the results of previous fuzzy-free research.