面向属性值遗漏数据决策树分类算法研究

Research on the Missing Attribute Value Data-oriented Decision Tree

在已有的多种决策树测试属性选择方法中,未见将属性值遗漏数据处理集成在测试属性选择过程中的报道,而现有的属性值遗漏数据处理方法都会不同程度地带入偏置。基于此,提出了一种将基于联合熵的信息增益率作为决策树测试属性选择标准的方法,用以在生成决策树的过程中消除值遗漏数据对测试属性选择的影响。在WEKA机器平台上进行了对比实验,结果表明,改进算法能够从总体上提高算法的执行效率和分类精度。

In the existing multiple choice methods of decision tree＇test attributes,can＇t see such report as ＂Let missing data processing integrated in the selection process of test attributes＂,however,the existing process methods of missing attribute value data could draw into bias in different degrees,based on this,proposed an information gain rate based on combination entropy as the decision tree＇s testing attributes selection criteria,which can eliminate missing value arrtibutes＇infulence on testing attributes selection,and carry out contrast experiments on WEKA.Experiment results indicate that the improvement can significantly increase whole efficiency and classification accuracy of the algorithm operation.