Margin & diversity based ordering ensemble pruning
Margin & diversity based ordering ensemble pruning
复制标题
利润
DOI：
10.1016/j.neucom.2017.06.052
复制
发表时间：
2018-01-31
期刊：
影响因子：
6
通讯作者：
Xu, Mingliang
中科院分区：
文献类型：
--
作者：
Guo, Huaping;Liu, Hongbing;Xu, Mingliang
集成剪枝是一种通过选择最优或次优子集作为最终集成进行预测来提高集成性能并减小集成大小的技术。在这项研究中，使用示例间隔和集成多样性，我们证明了集成修剪方法应该更多地关注以下两个因素：（1）具有小的绝对间隔的示例和（2）正确分类更多的示例和贡献更大的多样性的分类器。基于这一原则，我们提出了一种新的度量称为边缘和多样性为基础的措施（MDM）明确评估的重要性，个别分类。通过基于MDM以降序合并系综成员，形成子系综，使得用户可以选择前T个系综成员用于预测。与原始集成和其他最先进的集成剪枝方法相比，该方法在准确性方面表现出更好的性能。(C)2017爱思唯尔B.V.保留所有权利。
Ensemble pruning is a technique used to improve ensemble performance and reduce the ensemble size by selecting an optimal or sub-optimal subset as the final ensemble for prediction. In this research, using example margin and ensemble diversity, we prove that the ensemble pruning method should focus more on the following two factors: (1) examples with small absolute margin and (2) classifiers that correctly classify more examples and contribute larger diversity. Based on this principle, we propose a novel metric called the margin & diversity based measure (MDM) to explicitly evaluate the importance of individual classifiers. By incorporating ensemble members in a decreasing order based on the MDM, sub-ensembles are formed such that users can select the top T ensemble members for predictions. Compared to the original ensemble and other state-of`-the-art ensemble pruning methods, the proposed method shows better performance in terms of accuracy. (C) 2017 Elsevier B.V. All rights reserved.