Improving multi-label classification with missing labels by learning label-specific features
Improving multi-label classification with missing labels by learning label-specific features
复制标题
通过学习特定于标签的特征来改进缺失标签的多标签分类
DOI：
10.1016/j.ins.2019.04.021
复制
发表时间：
2019-08-01
影响因子：
8.1
通讯作者：
Huang, Qingming
中科院分区：
文献类型：
--
作者：
Huang, Jun;Qin, Feng;Huang, Qingming
现有的多标签学习方法主要利用由所有标签的所有特征组成的相同数据表示来区分所有标签，并假设每个训练样本都观察到所有类别标签。然而，在多标签学习中，每个类别标签可能由其自身的一些特定特征来确定，并且对于某些真实的应用，只能获得每个示例的部分标签集。本文提出了一种新的方法来学习标签特定的功能，多标签分类与缺失标签，命名为LSML。首先，一个新的补充标签矩阵是从不完整的标签矩阵，通过学习高阶标签相关性。然后，学习每个类别标签的标签特定的数据表示，并通过将学习到的高阶标签相关性结合起来，同时基于它构建多标签分类器。最后通过与现有方法的比较研究，验证了该方法的有效性。(C)2019爱思唯尔公司All rights reserved.
Existing multi-label learning approaches mainly utilize an identical data representation composed of all the features in the discrimination of all the labels, and assume that all the class labels are observed for each training sample. However, in multi-label learning, each class label might be determined by some specific features of its own, and only a partial label set of each example can be obtained for some real applications. This paper proposes a new method to learn Label-Specific features for multi-label classification with Missing Labels, named LSML. First, a new supplementary label matrix is augmented from the incomplete label matrix by learning high-order label correlations. Then, a label-specific data representation for each class label is learned, and the multi-label classifier is constructed simultaneously based on it by incorporating the learned high-order label correlations. A comparative study with the state-of-the-art approaches manifests the effectiveness of the proposed method. (C) 2019 Elsevier Inc. All rights reserved.