摘要
介绍了粗集理论的基本概念,通过对现有基于粗集的最小规则提取算法的分析,发现其缺陷,提出了一种改进的基于粗糙集的规则提取方法,并用UCI数据库中的9个标准数据集从规则集的规则数目、规则集的平均规则长度、规则集的平均规则支持、规则集的预测精度等4个指标对改进的算法进行了测试和对比分析,实验表明了该算法的有效性.
This paper introduces basic concepts of Rough Set Theory, finds its limitation by analyzing the current algorithm on inducing rules based Rough Set,proposes a improved algorithm on inducing rules based Rough Set,test and compares it with other algorithm in some aspects such as rules count of the rules set,average rule length of the rules set,average rule support of the rules set,forecast accuracy of the rules set used 9 standard datasets of UCI database. Experimentation makes sure validity of this new algorithm.
出处
《厦门大学学报（自然科学版）》
CAS
CSCD
北大核心
2004年第5期604-608,共5页
Journal of Xiamen University：Natural Science
基金
国家自然科学基金(10071074)
福建省教育委员会科技项目(JA022239)
厦门市重点科技项目(3502Z20031056)资助
