摘要
对知识发现中,概率统计方法与粗集理论方法在评价系统参数的重要性,分类隶属度与条件概率,输入条件与决策结果的依赖关系等方面进行了比较, 提出了将概率统计方法和粗集理论相结合,由确定性粗集的近似空间扩展到了不确定性的概率粗集的近似空间,为确定性和不确定性知识表达系统提供一个统计粗集模型。
The characteristics of knowledge discovery based on rough sets and statistical methods are compared on the significant factor in data attributes, class membership function and conditional probability, dependency relation of the decision consequents with respect to the input antecedents A combining approach of rough set with Bayesian methods is presented so that extends a determined rough knowledge representation space to an undetermined probabilistic approximate knowledge representation space, and suggests a probabilistic rough set model to realize determined knowledge manipulations and undetermined knowledge manipulations.
出处
《四川轻化工学院学报》
2003年第2期9-15,共7页
Journal of Sichuan Institute of Light Industry and Chemical Technology
基金
四川省应用基础研究项目02GY029-005基金的部分资助
关键词
数据挖掘
知识发现
粗集理论
概率统计
data mining
knowledge discovery
rough sets
statistical methods
