论文标题
半监督NER的基于局部添加性的数据增强
Local Additivity Based Data Augmentation for Semi-supervised NER
论文作者
论文摘要
命名实体识别（NER）是深度语言理解的第一个阶段之一，但当前的NER模型严重依赖于人类通知的数据。在这项工作中，为了减轻对标记数据的依赖性，我们为半监督NER提出了一种基于局部添加性的数据增强（LADA）方法，在该方法中，我们通过互相插值序列来创建虚拟样本。我们的方法有两个变体：lada和intrada，其中lada内部在一个句子中进行令牌之间的插值，而lada inser-lada样本则不同的句子以插值。通过采样培训数据之间的线性添加，LADA创建了无限量的标记数据，并改善了实体和上下文学习。我们通过为未标记的数据设计新颖的一致性损失，将LADA扩展到半监督的设置。在两个NER基准上进行的实验证明了我们方法对几个强基础的有效性。我们已在https://github.com/gt-salt/lada上公开发布了代码。
Named Entity Recognition (NER) is one of the first stages in deep language understanding yet current NER models heavily rely on human-annotated data. In this work, to alleviate the dependence on labeled data, we propose a Local Additivity based Data Augmentation (LADA) method for semi-supervised NER, in which we create virtual samples by interpolating sequences close to each other. Our approach has two variations: Intra-LADA and Inter-LADA, where Intra-LADA performs interpolations among tokens within one sentence, and Inter-LADA samples different sentences to interpolate. Through linear additions between sampled training data, LADA creates an infinite amount of labeled data and improves both entity and context learning. We further extend LADA to the semi-supervised setting by designing a novel consistency loss for unlabeled data. Experiments conducted on two NER benchmarks demonstrate the effectiveness of our methods over several strong baselines. We have publicly released our code at https://github.com/GT-SALT/LADA.
