MixLab:An Informative Semi-supervised Method for Multi-label Classification

摘要：

　　Multi-label classification is an intensively studied topic in data analysis.In spite of the considerable improvements,recent deep learning-based methods overlook the existence of unlabeled data,which consumes too much time on instance annotation.To circumvent this difficulty,semi-supervised multi-label classification aims to exploit the readily-available unlabeled data to help build multi-label classification model.To make full use of labeled and unlabeled data,this paper pro-pose a novel approach named MixLab,encourages the model classifi-cations to be accurate with label-correlated information and consistency regularization.It utilizes label correlations to enhance predicted labels for augmented unlabeled instances as targets and regularizes predictions to be consistent with this targets.We empirically validate the effective-ness of our framework by extensive experiments on four real datasets of textual content.

关键词： Semi-supervised learning Multi-label classification Text data augmentation

作者: Ye Qiu Xiaolong Gong Zhiyi Ma Xi Chen

作者单位: School of Electronics Engineering and Computer Science,Peking University,Beijing,China;Key Laborator Advanced Institute of Information Technology,Peking University,Hangzhou,China

会议类型: 国际会议

会议名称: 9th CCF International Conference on Natural Language Processing and Chinese Computing (NLPCC 2020)

会议地点: 郑州

会议语种:英文

页码: 506-518

在线出版日期: 2020-10-14（万方平台首次上网日期，不代表论文的发表时间）