Fuzzy Multiple Attribute Decision-Making via Association Rules Mining

With the advent of globalization, the social and economic environment has become complex and changeable, and the challenges faced by a decision-maker are full of uncertainty and ambiguity. Generally, the traditional decision-making method is too single and one-sided, and it will not have a great effect. As a result, researchers have begun to explore a variety of new decision-making methods. Among them, the association rule mining algorithm, which stands out, has been widely used in society. It has powerful data processing capabilities, can capture and process fuzzy information well, and reduces the omission of key decision-making factors. In this paper, FP-growth and triangular fuzzy number theory (FPGTFN) are applied to different datasets to verify the fuzzy decision theory. With multi-index and multi-index as important indicators, the fuzzy matrix iterates many times to calculate the highest weight. The optimal decision scheme is scheme 2, and the weight value is 0. 0756. The FPGTFN algorithm has better stability and less volatility in different datasets, and the decision results are more stable. FPGTFN has a smaller execution time, stable scheme output, and provides a better decision scheme. When the decision-maker has a certain expectation for the attribute values, the reference point can effectively avoid ignoring the correlation performance between the attributes, and can also reflect the actual profit and loss of the alternative scheme.