多尺度单值中智系统中基于优势粗糙集模型的最优尺度选择与约简

Optimal scale selection and reduction based on dominant rough set model in multi-scale single-valued neutrosophic systems

单值中智集是处理不确定、不一致信息的有效工具,结合单值中智粗糙集和多尺度决策系统,提出基于优势关系的多尺度单值中智粗糙集模型的最优尺度选择和约简算法.首先,在构建基于优势关系的多尺度单值中智粗糙集模型时引入正理想点、负理想点和不确定点来刻画单值中智数大小关系;其次,结合证据理论中的信任函数和似然函数给出模型的最优尺度选择算法及约简算法;最后,利用五组UCI数据集对文中提出的模型与算法进行实例验证,分析算法的有效性.提出的算法在分类精度和算法效率两方面都有所提高,进一步扩展了单值中智粗糙集在多尺度决策系统下的应用,为后续该领域的研究提供参考.

Single-valued neutrosophic sets are effective tools to deal with uncertain and inconsistent information. Combined with single-valued neutrosophic rough set and multi-scale decision systems,this paper proposes the optimal scale selection and reduction algorithms based on multi-scale single-valued neutrosophic dominance rough set model. First,when constructing multi-scale dominant single-valued neutrosophic rough set model,we use the ideal positive point,ideal negative point and most uncertain point to describe the dominance relationship between neutrosophic numbers. Second,combining with the belief function and plausibility function in evidence theory,we examine the optimal scale selection algorithm and reduction algorithm of the presented model. Third,we utilize five groups of UCI datasets to verify the model and algorithm proposed in this paper,and analyze the effectiveness of the algorithm. The algorithm proposed in this paper improves the classification accuracy and algorithm efficiency,and furtherly expands the application of single-valued neutrosophic rough set in multi-scale decision-making system,which provides a reference for subsequent research in this field.