摘要
海量数据属性约简的研究是数据挖掘研究中的一个难点。已有的许多属性约简算法对于空间复杂度考虑得不够,导致了算法不能适应大数据集的约简处理。结合分治法,在给定属性序下,提出了基于分治策略的属性约简算法。利用该算法可以快速得到海量数据的属性约简结果。仿真实验结果说明了该算法的高效性。
The attribute reduction of huge data is a difficult problem in the research of data mining. At present, many attribute reduction algorithms lack consideration on space complexity, which makes them cannot adapt to the reduction of large data set. In this paper, an attribute reduction algorithm of ordered attributes was proposed based on the divide and conquer, and this algorithm can be used to deal with huge data reduction. Simulation results show the efficiency of the algorithm.
出处
《重庆邮电大学学报（自然科学版）》
北大核心
2009年第4期455-460,共6页
Journal of Chongqing University of Posts and Telecommunications(Natural Science Edition)
基金
国家自然科学基金(60573068)
重庆市自然科学基金重点项目(2008BA2017)
重庆市教育委员会科学技术研究项目(KJ090512
KJ7150080050)
关键词
粗集
数据挖掘
分治
属性约简
属性序
rough set
data mining
divide and conquer
attribute reduction
attribute order
