Multi-Label Learning with Emerging New Labels
Multi-Label Learning with Emerging New Labels
复制标题
DOI：
10.1109/tkde.2018.2810872
复制
发表时间：
2018-10
影响因子：
8.9
通讯作者：
Yue Zhu;K. Ting;Zhi-Hua Zhou
中科院分区：
文献类型：
--
作者：
Yue Zhu;K. Ting;Zhi-Hua Zhou
在多标签学习任务中，一个对象拥有多个概念，每个概念由一个类标签表示。以往关于多标签学习的研究都集中在固定的类标签集上，即测试数据的类标签集与训练集中的类标签集相同。然而，在许多应用中，环境是动态的，数据流中可能会出现新概念。为了在这种环境中保持良好的预测性能，多标签学习方法必须具有检测和分类具有新标签的实例的能力。为此，我们提出了一种称为新标签多标签学习（MuENL）的新方法。它具有三个功能：对当前已知标签上的实例进行分类，检测新标签的出现，并为每个新标签构建一个与已知标签的分类器协作的新分类器。此外，我们还表明，通过简单地降低原始维度，然后将 MuENL 应用在降低的维度空间上，可以轻松扩展 MuENL 以处理稀疏高维数据流。我们的实证评估显示了 MuENL 在多个基准数据集上的有效性以及 MuENLHD 在稀疏高维微博数据集上的有效性。
In a multi-label learning task, an object possesses multiple concepts where each concept is represented by a class label. Previous studies on multi-label learning have focused on a fixed set of class labels, i.e., the class label set of test data is the same as that in the training set. In many applications, however, the environment is dynamic and new concepts may emerge in a data stream. In order to maintain a good predictive performance in this environment, a multi-label learning method must have the ability to detect and classify instances with emerging new labels. To this end, we propose a new approach called Multi-label learning with Emerging New Labels (MuENL). It has three functions: classify instances on currently known labels, detect the emergence of a new label, and construct a new classifier for each new label that works collaboratively with the classifier for known labels. In addition, we show that MuENL can be easily extended to handle sparse high dimensional data streams by simply reducing the original dimensionality, and then applying MuENL on the reduced dimensional space. Our empirical evaluation shows the effectiveness of MuENL on several benchmark datasets and MuENLHD on the sparse high dimensional Weibo dataset.