概念漂移探测是数据流挖掘的一个研究重点，不确定性分析是粗糙集理论的研究核心之一。结合数据流、概念漂移和粗糙集、F‐粗糙集的基本观点，以上下近似为工具，定义了上下近似概念漂移、上下近似概念耦合等概念，据此分析了信息表内概念随着属性而变化的特点。以正区域为工具，定义了决策表内概念漂移、概念耦合等概念，分析了决策表内整体概念随属性变化而变化。在认识论方面，从理想和现实2方面定义了认识收敛，从粒计算、粗糙集的角度对人类认识世界的方式进行了探讨。%Concept drifting detection is one of hot topics in data stream mining , and analysis of uncertainty is dominant in rough set theory . Combined with the ideas of data stream , concept drifting , rough sets and F‐rough sets , a lot of concepts such as concept drifting of upper approximation ,concept drifting of lower approximation ,concept coupling of upper approximation and concept coupling of lower approximation etc are defined .The change of concepts in an information system is analyzed with these definitions .With the positive region ,integral concept drifting ,integral concept coupling are defined .The analysis and measurement for the change of concept uncertainty are conducted .From the view of epistemology ,the concept of cognition convergence is defined from the ways of idealism and realism .It provides heuristic information for realizing the world of human beings from the viewpoints of granular computing and rough sets .
展开▼
