Outlier Detection in Single Universal Set using Intuitionistic Fuzzy Proximity Relation based on A Rough Entropy-Based Weighted Density Method

Data mining is a technique for analyzing larger datasets to identify patterns, information, and relationships that may be used to solve challenging problems. Identifying outliers has attracted the focus of researchers working on a variety of areas. Outliers are things that behave differently from other objects. With real-world data, rough set theory can cope with ambiguity and uncertainty. So far, the study has solely focused on spotting outliers using the membership function. Outliers may be recognized using membership and non-membership values, however, utilizing the principle of intuitionistic fuzzy proximity relation. At this step, the indiscernibility of objects is discovered, and the quantitative data is then converted to qualitative data. This article proposes outlier detection in single universal sets using an intuitionistic fuzzy proximity relation with a rough set based on complement entropy and weighted density approach. The empirical study has been considered for ranking the colleges based on the parameters evaluated.