An adaptive semi-supervised ensemble method, ASSEMBLE, is proposed that constructs classification ensembles based on both labeled and unlabeled data. ASSEMBLE alternates between assigning "pseudo-classes" to the unlabeled data using the existing ensemble and constructing the next base classifier using both the labeled and pseudo-labeled data. Mathematically, this intuitive algorithm corresponds to maximizing the classification margin in hypothesis space as measured on both the labeled and unlabeled data. Unlike alternative approaches, ASSEMBLE does not require a semi-supervised learning method for the base classifier. ASSEMBLE can be used in conjunction with any cost-sensitive classification algorithm for both two-class and multi-class problems. ASSEMBLE using decision trees won the NIPS 2001 Unlabeled Data Competition. In addition, strong results on several benchmark datasets using both decision trees and neural networks support the proposed method.
展开▼
机译：提出了一种自适应半监督集成方法ASSEMBLE，该方法基于标记和未标记的数据构造分类集合。 ASSEMBLE在以下两种方法之间进行交替：使用现有的集合为未标记的数据分配“伪类”，以及使用标记的数据和伪标记的数据构造下一个基本分类器。在数学上，此直观算法对应于最大化在标记和未标记数据上测得的假设空间中的分类裕度。与替代方法不同，ASSEMBLE不需要针对基本分类器的半监督学习方法。 ASSEMBLE可以与任何成本敏感的分类算法结合使用，以解决两类和多类问题。大会使用决策树赢得了NIPS 2001 Unlabeled Data Competition。另外，在使用决策树和神经网络的几个基准数据集上的强大结果支持了该方法。
展开▼
