21 / 12 / 2022 - 13:30 – 14:00

Safiye Turgay

Sakarya University, Turkey

Rough-Set-Based Decision Model and MCDM Applications for Incomplete Information Systems

Abstract

Today, the enormous increase and the complexity of the data make it inevitable to use rough set in the data analysis process. With the increasing data traffic, uncertainty and deficiency situations require ordering and grouping data variables according to their importance and data quality before the learning process. In this study, the analysis structure of decision models and MCDM methods was discussed in the process of data analysis. Increasing the pre-learning data quality of the features will also positively affect the quality and speed of learning. The rough set approach includes the advantages of obtaining information directly, keeping the information in mind, reducing information and evaluating the relationship between variables through reasoning. It understands the classification as the equivalence relation in a particular space, and the equivalence relation constitutes the division of the space. In this study, the main features of coarse clustering and the role it played in the data analysis process were emphasized. Data analysis and multi-criteria decision making technique are among the complementary subjects of decision science. In this context, the pre-analysis quality of the existing data is increased with rough cluster-based number analysis. Data analysis, data mining and MCDM methods were handled as a whole. Especially with the analysis of incomplete data, it is aimed to select the rules that will help decision-making. The main differences between them were examined by examining the studies done so far.