基于二进制有序差别集的属性约简算法

AN ATTRIBUTE REDUCTION ALGORITHM BASED ON BINARY ORDERED DISCERNIBILITY SET

文献[6]给出的基于简化二进制可分辨矩阵的快速属性约简算法是不完备的,并且在处理大数据集时的效率不很理想。提出一种基于二进制有序差别集的属性约简算法,该算法不需要创建二进制可分辨矩阵,减少了数据处理量,大大提高了约简的效率,使算法的时间复杂度和空间复杂度分别降为m ax{O(|C|2|U/C|2),O(|C|2|BM sCount|)}和O(|BM sCount|)。最后的实验结果表明该算法是正确的、高效的。

Fast attributes reduction algorithm provided by Xu Yanzhang based on simplified binary discernibility matrix is an incomplete one, and is far from efficiency in dealing with large data sets. In the paper, an attribute reduction algorithm based on binary ordered discern- ibility set is presented. The algorithm does not need to create a binary discernibility matrix, and the computing data is decreased greatly, thus the efficiency of the reduction algorithm is noticeably improved. The time complexity and space complexity of the new algorithm are cut down to max{O（｜C｜U/CI2）,0（｜C｜2}BMsCount1）｜ and O（IBMs Count｜）.. The final experimental results show that the algorithm is correct and efficient.