arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2008.06695cs.CL
收藏

用于多标签文本分类的全标签文档预训练

Label-Wise Document Pre-Training for Multi-Label Text Classification

Center of Intelligence Science and Technology, Beijing University of Posts and Telecommunications(北京邮电大学智能科学与技术中心)

Han Liu, Caixia Yuan, Xiaojie Wang

摘要:

多标签文本分类(MLTC)的一个主要挑战是刺激性地利用可能的标签差异和标签相关性。在本文中,我们通过开发Label-Wise Pre-Training(LW-PT)方法来解决这一挑战,以获得具有标签感知信息的文档表示。其基本思想是,多标签文档可以表示为多个标签表示的组合,并且相关标签总是在相同或相似的文档中共存。LW-PT通过构建标签式文档分类任务和训练标签式文档编码器来实现这一思想。最后,预训练的标签式编码器通过下游MLTC任务进行微调。大量的实验结果表明,该方法具有显着的优势,比以前的国家的最先进的模型,能够发现合理的标签关系。该代码的发布是为了方便其他研究人员。

英文摘要:

A major challenge of multi-label text classification (MLTC) is to stimulatingly exploit possible label differences and label correlations. In this paper, we tackle this challenge by developing Label-Wise Pre-Training (LW-PT) method to get a document representation with label-aware information. The basic idea is that, a multi-label document can be represented as a combination of multiple label-wise representations, and that, correlated labels always cooccur in the same or similar documents. LW-PT implements this idea by constructing label-wise document classification tasks and trains label-wise document encoders. Finally, the pre-trained label-wise encoder is fine-tuned with the downstream MLTC task. Extensive experimental results validate that the proposed method has significant advantages over the previous state-of-the-art models and is able to discover reasonable label relationship. The code is released to facilitate other researchers.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑