A Modified Sugeno Integral Based Similarity Measure for Intuitionistic Fuzzy Soft Set and its Application

In the paper, intuitionistic fuzzy soft set (IFSS) is considered for the mathematical modeling of real-world issues. The IFSS involves an extra parameter to extrapolate the traditional notion of intuitionistic fuzzy set, and hence it is highly utilized for real-life applications. The similarity measures for IFSSs are defined to deal with the problems of pattern recognition, machine learning, and decision-making. This paper introduces a Sugeno integral based similarity measure for IFSSs in which a novel concept of
$(\alpha,\beta)-cut$
is applied for the first time. Sugeno integral is a measure to calculate the total expected value of similarity between two IFSSs. Since IFSS consists of two independent components,
$(\alpha,\beta)-cut$
is used to ensure the operationalization of the measure. Here, some of the mathematical properties related to the proposed measure are also studied. Moreover, we compare the performance of the proposed measure with some well-known similarity measures over a benchmark example. We have further validated the efficiency of the proposed measure on the basis of accuracy, sensitivity, F-measure, and error rate over UCI Machine Learning Repository data set.