计算机应用与软件2016，Vol.33Issue(9)：264-268,295,6.DOI:10.3969/j.issn.1000-386x.2016.09.062
基于测试代价敏感的不完备决策系统属性约简算法
ATTRIBUTE REDUCTION ALGORITHM OF INCOMPLETE DECISION SYSTEM BASED ON TEST COST SENSITIVITY
摘要
Abstract
We introduced the problem of test-cost-sensitive attribute reduction in incomplete decision system,and suggested the definition of inconsistent object set and an algorithm for computing the inconsistent object set.According to the nature of inconsistent object set we improved the definition of attribute significance.Considering the test cost factors and the varied amount of the number of inconsistent objects we presented a new definition of attribute significance and the weight setting method of it.And then we gave the calculation algorithm of attribute significance. Based on these conditions, we proposed a heuristic attribute reduction algorithm with the time complexity O(k |C|2 |U |)and the space complexity O(|U |).Through theoretical analysis,example analysis and experiment analysis we explained the accuracy and feasibility of the reduction algorithm.关键词
测试代价敏感/不完备决策系统/属性重要性/属性约简/不一致对象Key words
Test-cost-sensitive/Incomplete decision system/Attribute significance/Attribute reduction/Inconsistent object分类
信息技术与安全科学引用本文复制引用
谢小军,徐章艳,乔丽娟,朱金虎..基于测试代价敏感的不完备决策系统属性约简算法[J].计算机应用与软件,2016,33(9):264-268,295,6.基金项目
国家自然科学基金项目（61262004，61363034，60963008）；广西自然科学基金项目（2011GXNSFA018163）；八桂学者专项基金。 （）
