摘要
当标注样本匮乏时,半监督学习利用大量未标注样本解决标注瓶颈的问题,但由于未标注样本和标注样本来自不同领域,可能造成未标注样本存在质量问题,使得模型的泛化能力变差,导致分类精度下降.为此,基于wordMixup方法,提出针对未标注样本进行数据增强的u-wordMixup方法,结合一致性训练框架和Mean Teacher模型,提出一种基于u-wordMixup的半监督深度学习模型(semi-supervised deep learning model based on u-wordMixup,SD-uwM).该模型利用u-wordMixup方法对未标注样本进行数据增强,在有监督交叉熵和无监督一致性损失的约束下,能够提高未标注样本质量,减少过度拟合.在AGNews、THUCNews和20 Newsgroups数据集上的对比实验结果表明,所提出方法能够提高模型的泛化能力,同时有效提高时间性能.
When labeled data are deficient,semi-supervised learning uses a large number of unlabeled data to solve the bottleneck problem of labeled data.However,as the unlabeled data and labeled data come from different fields,quality problems of unlabeled data would be callsed,which makes the generalization ability of the model poor and leads to the degradation of classification accuracy.Therefore,based on the wordMixup method,this paper proposes the u-wordMixup method for data augmentation of unlabeled data,and a semi-supervised deep learning model based on the u-wordMixup(SD-uwM)by combining the consistent training framework and the Mean Teacher model.The model utilizes the u-wordMixup method to augment the data of unlabeled data,which can improve the quality of unlabeled data and reduce overfitting under the constraints of supervised cross-entropy and unsupervised consistency loss.The comparative experimental results on the datasets of AGNews,THUCNews and 20 Newsgroups show that the proposed method can improve the generalization ability of the model and also effectively improve the time performance.
作者
唐焕玲
宋双梅
刘孝炎
窦全胜
鲁明羽
TANG Huan-ling;SONG Shuang-mei;LIU Xiao-yan;DOU Quan-sheng;LU Ming-yu(School of Computer Science and Technology,Shandong Technology and Business University,Yantai 264005,China;School of Information and Electronic Engineering,Shandong Technology and Business University,Yantai 264005,China;Co-innovation Center of Shandong Colleges and Universities:Future Intelligent Computing,Yantai 264005,China;Key Laboratory of Intelligent Information Processing in Universities of Shandong,Shandong Technology and Business University,Yantai 264005,China;Information Science and Technology College,Dalian Maritime University,Dalian 116026,China)
出处
《控制与决策》
EI
CSCD
北大核心
2023年第6期1646-1652,共7页
Control and Decision
基金
国家自然科学基金项目(61976124,61976125,62176140,61873177,61972235,82001775).
关键词
半监督学习
数据增强
深度学习
文本分类
semi-supervised learning
data augmentation
deep learning
text categorization