THE CONCEPT LEARNING IN THE THEORY OF ROUGH SETS

摘要：

Knowledge reduction in decision table is important in both theory and application, and it outputs a minimal algorithm as a result. Set of the samples fitting the minimal algorithm is a concept over the set of all possible instances. But in unfamiliar environment, decision table is obtained randomly. So the obtained concept is an approximation to a potential target concept. We discuss the model of this concept learning, sample complexity of its hypothesis space and PAC-learnability of its target concept class.

关键词： Rough Set Concept Learning Sample Complezity PAC-Learnability

作者: QUN-FENG ZHANG YU-TING JIANG ZHI-QIANG LI

作者单位: Key Lab.of Machine Learning and Computational Intelligence, College of Mathematics and Computer Scie Industrial and commercial college, Heibei University, Baoding, 071002, China Hebei Information Engineering School, Baoding, 071000, China

会议类型: 国际会议

会议名称: 2008 International Conference on Machine Learning and Cybernetics(2008机器学习与控制论国际会议)

会议地点: 昆明

会议语种:英文

页码: 337-339

在线出版日期: 2008-07-12（万方平台首次上网日期，不代表论文的发表时间）