一种可调阈值的数据最优尺度选择算法

A Data Optimal Scale Selection Algorithm with Adjustable Threshold

粒度计算正向多粒度层次选择及数据最优尺度选择等方向发展。对具有不同尺度等级的数据组合进行相关研究。针对具有不同数据尺度的系统,构建了一种可调阈值的数据最优尺度选择算法,以便选择数据中的最优尺度,从而获得最优尺度组合。最后,结合决策树分类算法,并采用5个UCI公开数据集实验验证了所提出算法的可行性和有效性。

Granular computing is developing in the direction of multi-granularity selection and data optimal scale selection.In the paper,the authors conduct related research on data combin-ations with different scale levels.For the systems with different data scales,a data optimal scale selection algorithm with adjustable threshold is constructed to select the optimal scale in the data and obtain the optimal scale combination.Finally,combined with the decision tree classification algorithm,and used five UCI public dataset,the feasibility and effectiveness of the proposed algorithm are verified.