!!具有全局聚类多属性离散化算法_刘弹.pdfVIP

45 9 西 安 交 通 大 学 学 报 Vol45 No9 2011 9 JOURNAL OF XIAN JIAOT ONG UNIV RSIT Y Sep. 2011 , , ( , 710049, ) : 为了减少连续属性离散化后有用信息的丢失和信息系统总的断 数量, 提出了一种具有全 局聚类效果的多属性离散化算法. 算法根据各属性预插入断 对信息系统近似分类质量的影响, 来 确定要插入断 的属性, 从全局属性范围选择最佳断 . 根据 Ameva 统计量来判断属性中最佳断 的位置, 并以保证决策表的近似分类质量作为算法的终止条件. 实验采用多组机器学习数据对算 法的性能进行了检验, 并与几种经典算法做了对比. 实验结果表明, 用新的离散化算法获得的结果 所建的C45 决策树分类模型, 具有较好的分类精度和较少的节 数量. : 统计量; 连续属性; 离散化 : TP18 : A : 0253-987X(2011) 09-0001-05 Synchronized Continuous Attributes Discretization Based on Ameva LIU Dan, YANG Jingming, LUO Ailing ( State Key Laboratory for Manufacturing Systems ngineering, Xian Jiaotong University, Xian 710049, China) Abstract: To avoid information loss and cut points decrease after discretization of continuous at- tributes, a synchronized continuous attribute discretization algorithm with good global clustering effect for selecting cut points from all conditions attributes is presented. T his algorithm decides which continuous attribute should be inserted according to the cut point from all attributes based on the influence of the inserted cut point. T he influence is evaluated by information system ap- proximation classification quality. T hen cut point is selected from the candidate points in the at- tribute according to Ameva statistics, and the level of indiscernibility relation is chosen as the stopping condition of the algorithm. By UCI machine learning data sets a comparison with several classic discretization algorithms shows that the C45 classification model based on the proposed al- gorithm is of good classification accuracy and needs less nodes. Keywords: statistics; continuous attributes; discretization , , , [ 1]