基于变精度粗集和搜索树的增量规则获取算法  

Incremental rule-extraction algorithm based on variable precision rough set and search tree

作　　者：邱兆雷[1] 王爱云[1] 陈传臻[2] 

机构地区：[1]山东师范大学管理与经济学院,济南250014 [2]山东师范大学信息科学与工程学院,济南250014

出　　处：《计算机工程与应用》2008年第14期163-165,共3页Computer Engineering and Applications

基　　金：泰山学者专项建设经费

摘　　要：基于可变精度粗糙集模型和搜索树提出了一种新的增量式规则获取算法。该算法引入可变精度粗糙集模型以已获取规则集为启发信息,通过对解空间进行深度优先启发式搜索产生新的不确定性规则;并通过对原有规则置信度的更新,给出了原有规则集的更新算法;最后给出了实例分析。The incremental rules extraction is a focus problem of KDD. A novel rule extraction algorithm which is called ＂RDBVPRST＂ （Rule Derivation Based on Variable Precision Rough Set and Search Tree） is proposed. It is one kind of depth first heuristic search algorithms. Based on this algorithm, the new rules are extracted and the current rule set is updated by updating the rules＇ confidence degree. At last, an example is given to illustrate the characteristics of this new incremental algorithm.

关 键 词：可变精度粗糙集 规则获取 搜索树 深度优先启发式搜索算法 

分 类 号：TP182[自动化与计算机技术—控制理论与控制工程]