动态多尺度决策信息系统局部最优尺度的更新规律

Updating Law of Local Optimal Scale of Dynamic Multi-Scale Decision Information System

在对象动态增加的情况下,对多尺度决策信息系统(MDIS)保持局部决策类不确定性的最优尺度更新规律进行研究。首先,介绍决策信息系统和多尺度决策信息系统决策类不确定性的基本知识,以及MDIS保持局部决策类不确定性的最优尺度定义。然后,在增加一个对象的条件下,分析MDIS局部决策类不确定性的更新规律。最后,采用增量学习方法,给出增加一个对象条件下MDIS局部最优尺度不变和变大的充分必要条件。结果表明:文中方法可以快速地确定更新系统局部最优尺度。

Research on the updating law of optimal scale of multi-scale decision information system(MDIS)to keep local decision class uncertainty under the condition of dynamic increase of objects.Firstly,the basic knowledge of decision class uncertainty of decision information system and multi-scale decision information system are introduced,and the definition of optimal scale of MDIS to keep local decision class uncertainty is given.Then,the updating law of local decision class uncertainty of MDIS is analyzed under the condition of adding one object.Finally,using the incremental learning method,the sufficient and necessary conditions are given for the local optimal scale of MDIS to remain invariant or increase under the condition of adding an object.The results show that the proposed method can quickly determine the local optimal scale of the update system.