A new decision-theoretic rough set model and minimum decision cost attribute reduction algorithm

There are a lot of numerical data in practical application, but traditional decision-theoretic rough sets can only deal with symbolic data. In order to improve this limitation, this paper constructs a fuzzy neighborhood decision-theoretic rough set model and proposes a minimum decision cost attribute reduction algorithm. Firstly, the fuzzy rough set and neighborhood rough set are integrated into decision-theoretic rough set, and the fuzzy neighborhood decision-theoretic rough set is proposed, which makes the model have the advantages of both fuzzy rough set and neighborhood rough set in dealing with numerical data. Then, based on the model, a definition of decision cost is given and the corresponding minimum decision cost attribute reduction is proposed. Finally, the experimental results show that the proposed algorithm has better cost-sensitive attribute reduction performance.