The Possible Equivalent Value Set for Incomplete Data Set.

Incomplete or missing data is a significant challenge in real-world information systems that can lead to flawed decision-making. The rough set theory has limitations when dealing with incomplete information systems, and researchers have proposed alternative approaches. This paper proposes a new similarity relation that utilises probable equivalent value sets to improve the accuracy of incomplete information systems. The approach enhances the quality of decision-making and provides reliable results. Experiments conducted on various datasets with different levels of missing data show that the approach can improve the accuracy of incomplete information processing by up to 90%. Compared to existing methods, the proposed approach can handle both categorical and continuous attributes, address problems of non-uniqueness and redundancy of probable equivalent value sets, and is not dependent on any specific data distribution. In conclusion, the proposed technique provides an effective solution to the challenges of incomplete information systems. By using probable equivalent value sets, the approach improves the accuracy of incomplete information processing and enhances the quality of decision-making. It has potential for applications in data science and related areas, and further research is needed to explore its limitations and applicability to real-world problems.