论文部分内容阅读
针对粗糙集理论中的属性约简问题做了探讨研究。从寻找属性约简的角度,首先描述了决策表中的属性的重要性,并利用已求得的正区域使处理数据的范围不断缩小,约简集中的属性从核集开始,通过向属性核添加重要性最大的属性,得到属性的最小相对约简。从而减少求约简的时间。最后进行实证,该算法同传统的算法相比,在计算量减少的同时能得到更简约的结果,证明了该算法的正确性和可行性。
Aiming at the problem of attribute reduction in rough set theory, From the perspective of finding the attribute reduction, firstly, the importance of the attributes in the decision table is described, and the range of the processed data is continuously reduced by using the obtained positive region. Attributes in the reduced set begin from the kernel set, Add the most important attributes, get the minimum relative reduction of attributes. Thus reducing the time to seek reduction. Finally, an empirical test is carried out. Compared with the traditional algorithm, this algorithm can obtain more simple results while reducing the computational load, and proves the correctness and feasibility of the algorithm.