基于区分对象对的不完备决策表求核

Computing core based on discernibility object pair set in incomplete decision table

在差别矩阵的基础上,针对不完备决策表提出了基于差别矩阵的区分对象对集定义,并证明求不完备决策表的核可以转化到求基于差别矩阵的区分对象对集上。在此基础上,提出了一种基于区分对象对的不完备决策表求核算法,该算法的时间复杂度为:max{O(|C||U||U pos|)O(K|C||U|)},优于同类算法的时间复杂度;用实例说明了新算法的有效性。

The definition of discernibility object pair set of incomplete decision table, based on discernibility matrix, is defined. And it is proved that computing the core of incomplete decision table is equal to computing the discernibility object pair set of incomplete decision table. Then an algorithm for computing core based on discernibility object pair set of incomplete decision table is proposed. And the time complexity of the new algorithm is max{O（K｜C｜U｜）、O（｜C｜U｜Upos｜）}, which is better than the time complexity of the same kind of algorithms. At last, an example is used to illustrate the efficiency of the new algorithm.