Multi-criteria Attribute Ranking Based on Rough Sets and Interval-Valued Fuzzy Sets
摘要
The paper presents an application of the Multi Criteria Decision Making (MCDM) technique based on incomplete Interval-valued Fuzzy Preference Relation (IV-FPR) to the attribute ranking problem. The concept involves modifying existing methods within the RAFAR (Rough-fuzzy Algorithm for Attribute Ranking) framework, where multiple IV-FPRs are used to express preferences between attributes, with each corresponding to a specific decision class. In addition, the ranking method that converts a set of IV-FPR into a single weight vector will be discussed. The proposed method is specifically designed for decision tables with multiple decision classes and it will be illustrated using benchmark datasets.