论文部分内容阅读
等价类是Rough集理论的核心概念之一,如何高效地计算等价类是提高各相关算法性能的关键.引入高维空间向量夹角的概念,根据数据在机器中的存储特点,以字节内容的最大值加1作为基数对数据进行基数排序,在此基础上设计以计算向量夹角来求信息系统等价类的算法.该算法把原来计算等价类的逻辑比较转换为数值计算,非常显著地提高了等价类的计算效率,尤其对大规模高维数据.该算法的时间复杂度为O(|C‖U|log|U|),理论分析与实验结果表明了该算法的正确性和高效性.
Equivalence classes are one of the core concepts in Rough set theory. How to compute equivalence classes efficiently is the key to improve the performance of the algorithms.According to the concept of vector angle in high-dimensional space, The contents of the maximum plus 1 as the base of the data base sort, on the basis of the design to calculate the angle between the vector information system equivalence class algorithm.This algorithm to calculate the original class of equivalent logic equivalent conversion to numerical calculation , Which improves the computational efficiency of equivalence classes significantly, especially for large-scale and high-dimensional data.The time complexity of the algorithm is O (| C | U | log | U |), and the theoretical analysis and experimental results show that the algorithm The correctness and efficiency.