Three-way Decision Movement Strategy Based on Hierarchical Clustering

Abstract

Acting is an important step in three-way decision TAO model,and it is also an important method to realize object movement.By implementing the strategies,the object is moved from a disadvantageous area to the advantageous area.In recent years,scholars have proposed two kinds of movement strategies,one is region-based movement,the other is object-based movement.However,the two movement strategies are analyzed and formulated from a single-level perspective,and the formulation of movement strategy is not considered from a multi-level perspective.Therefore,in order to make multi-level movement strategy,this paper introduces hierarchical clustering and proposes a three-way decision movement strategy based on hierarchical clustering.Firstly,it uses hierarchical clustering to divide the objects in the disadvantageous area into different levels,and the clustering results are different at each level.Then,according to the highest frequency of global attribute value criterion,a movement strategy is formulated for clusters in each hierarchy,and different clusters have different movement strategies.In addition,the paper also uses the benefit and cost of the movement process to evaluate the different levels of the movement strategies.Finally,experimental results prove the validity of the proposed model.