论文部分内容阅读
将粗糙集理论与模糊集理论结合起来,给出一种连续值域决策表的离散化算法。该算法从已知数据的初始决策系统出发,首先构造对像的相似矩阵,然后根据相似矩阵的传递闭包及粗糙集正域的思想得出决策表的条件类,再根据条件类将连续值决策表化为区间值决策表,最后根据各区间值将连续值域决策表化为离散决策表。
The rough set theory and fuzzy set theory are combined to give a discretization algorithm of continuous range decision table. The algorithm starts with the initial data-aware decision system. Firstly, the similarity matrix of the image is constructed, and then the conditional classes of the decision table are derived according to the idea of the transitive closure of the similar matrix and the positive domain of the rough set, and then the continuous values The decision table is transformed into the interval value decision table, and finally the continuous range decision table is transformed into the discrete decision table according to each interval value.