基于相关系数加权朴素信念分类模型

Weighted naive credal classification model based on correlation coefficients

针对目前大多数分类器简单抛弃缺失数据的问题,基于朴素信念分类提出了一种有缺失值实例的加权保守推理规则的分类算法。以数据集特征属性与决策属性之间的相关系数作为权值,根据有缺失值实例加权保守推理规则对有非随机缺失属性的待分类实例所有可能的类别进行选择。实验结果表明,提出的基于有缺失值实例的加权保守推理规则分类算法有效地提高了分类性能,是一种有效的缺失数据集分类算法。

Aimed at the ignorance arising from missing data, the weighted conservative inference rule classifier based on correlation coefficients is proposed, in which the prediction class is selected form a set of possible classes according to the weighted conservative inference rule. The weights are the corresponding correlation coefficients between the feature attributes and the decision attribute. Simulation results of a variety of UCI data sets verify the effectiveness of this algorithm.