MAGDM-Miner

Abstract

In this article, the authors propose a new framework called the MAGDM-Miner, for mining correlation rules from trapezoidal intuitionistic fuzzy data efficiently. In the MAGDM-Miner, the raw data from a Multiple Attribute Group Decision Making (MAGDM) problem with trapezoidal intuitionistic fuzzy data are first pre-processed using some arithmetic aggregation operators. The aggregated data in turn are processed for efficient data selection through fuzzy correlation rule mining where the unwanted or less important decision variables are pruned from the decision making system. Using this MAGDM-Miner, a decision-maker can overcome the drawbacks in the conventional methods of Decision Support Systems (DSS) especially when dealing with large data-set. The algorithm is also presented, in which the technique of Fuzzy Correlation Rule Mining (FCRM) is fused into the MAGDM problem, in order to enhance the efficiency and accuracy in decision making environment. A numerical illustration is presented to show the effectiveness and accuracy of the newly developed MAGDM-Miner algorithm.