在分析贝叶斯粗糙集模型已有的约简算法的基础上,从含有多个决策类情况下的全局相对增益函数的角度,利用二进制编码方法给出一种求贝叶斯粗糙集所有约简及核的算法,并基于实际应用,给出对求出的所有约简进行筛选的拓展算法。通过算例说明算法的实用性和有效性。
With an analysis based on previous method of redution algorithm of bayesian rough sets modle, from the view of the global relative gain function contained of multiple decision classes,based on the binary code method, an algorithm for solving all reductions and nuclear is given. Furthermore,the proposed algorithm is extended from the point of view of the practical appliction, a method to screening all of the reduction is propsed, and the output is the optimal reduction. Finally,an example is used to demonstrate that the proposed algorithm is practical and effective.