摘要
在数据挖掘和机器学习的过程中,分类器的主要任务是提高数据分类的精确度和降低数据分类的费用开销,本文针对传统分类器只考虑了如何提高数据分类的精确度而没有考虑到如何降低数据分类的开销缺陷,提出了一种基于属性开销约束的矩阵属性约简算法,定义了一种新的函数作为属性约简的启发信息,探讨了基于矩阵方法计算等价关系矩阵的增量更新机制.该算法缩短了粗糙集属性约简的计算时间,保证了属性约简的实时性,并通过实例进一步验证了所提出方法的有效性和正确性.
In many data mining and machine learning applications,there are two objectives in the task of classification. One is decreasing the test cost,the other is improving the classification accuracy,most existing research work focuses on the latter. In this paper,we point out that test cost must be under taken in parallel,attribute reduction is mandatory in dealing with the former objective. With this in mind,a new information function is designed,where function are decided by test costs and attribute significance. Then,the matrix-based incremental mechanisms for equivalence relation matrix are discussed,and a matrix attribute reduction approach with test cost constraint is proposed. The proposed attribute reduction algorithm can find a reduct in a short time. Our example illustrate that the algorithm is effective.
作者
罗爱玲
景运革
LUO Ai-ling;JING Yun-ge(Mechanical and Electrical Engineering Department,Yuncheng College,Yuneheng 044000,Shanxi,China;Department of Public Computer Teaching,Yuncheng College,Yuncheng 044000,Shanxi,China)
出处
《山西师范大学学报（自然科学版）》
2018年第3期21-26,共6页
Journal of Shanxi Normal University(Natural Science Edition)
基金
国家自然科学基金面上项目(61573292)
运城学院院级项目(YQ-2017028)
关键词
粗糙集
属性约束
正域约简
关系矩阵
增量机制
rough set
test cost constraint
positive domain reduction
relation matrix
incremental mecha-nisms
