期刊文献+

基于句式元学习的Twitter分类 预览

Sentence Style Meta Learning for Twitter Classification
在线阅读 免费下载
分享 导出
摘要 针对多类别的社交媒体短文本分类准确率较低问题,提出一种学习多种句式的元学习方法,用于改善Twitter文本分类性能。将Twitter文本聚类为多种句式,各句式结合原类标签,成为多样化的新类别,从而原分类问题转化为较多类别的few-shot学习问题,并通过训练深层网络来学习句式原型编码。用多个三分类Twitter数据来检验所提Meta-CNN方法,结果显示,该方法的学习策略简单有效,即便在样本数量不多的情况下,与传统机器学习分类器和部分深度学习分类方法相比,Meta-CNN仍能获得较好的分类准确率和较高的F1值。 Due to the limited length and freely constructed sentence structures,it is a difficult classification task for short text classification,especially in multi-class classification.An efficient meta learning framework is proposed for twitter classification.The tweets are clustered into many sentence styles corresponding to new class labels.Thus,the original text classification task becomes few-shot learning task.When applying few-shot learning on benchmark datasets,the proposed method Meta-CNN achieves improvement in accuracy and F1 scores on multi-class twitter classification,and outweigh some traditional machine learning methods and a few deep learning approaches.
作者 闫雷鸣 严璐绮 王超智 贺嘉会 吴宏煜 YAN Leiming;YAN Luqi;WANG Chaozhi;HE Jiahui;WU Hongyu(School of Computer and Software&Jiangsu Engineering Center of Network Monitoring,Nanjing University of Information Science and Technology,Nanjing 210044)
出处 《北京大学学报:自然科学版》 CAS CSCD 北大核心 2019年第1期98-104,共7页 Acta Scientiarum Naturalium Universitatis
基金 国家自然科学基金(61772281,61703212,61602254)资助.
关键词 元学习 少次学习 情感分析 卷积神经网络 meta learning few-shot learning sentiment analysis CNN
作者简介 通信作者:闫雷鸣,E-mail:lmyan@nuist.edu.cn
  • 相关文献

参考文献2

二级参考文献85

  • 1Ben-David S,Blitzer J,Crammer K,Pereira F.Analysis of representations for domain adaptation.In:Platt JC,Koller D,Singer Y,Roweis ST,eds.Proc.of the Advances in Neural Information Processing Systems 19.Cambridge:MIT Press,2007.137-144. 被引量:1
  • 2Blitzer J,McDonald R,Pereira F.Domain adaptation with structural correspondence learning.In:Jurafsky D,Gaussier E,eds.Proc.of the Int’l Conf.on Empirical Methods in Natural Language Processing.Stroudsburg PA:ACL,2006.120-128. 被引量:1
  • 3Dai WY,Xue GR,Yang Q,Yu Y.Co-Clustering based classification for out-of-domain documents.In:Proc.of the 13th ACM Int’l Conf.on Knowledge Discovery and Data Mining.New York:ACM Press,2007.210-219.[doi:10.1145/1281192.1281218]. 被引量:1
  • 4Dai WY,Xue GR,Yang Q,Yu Y.Transferring naive Bayes classifiers for text classification.In:Proc.of the 22nd Conf.on Artificial Intelligence.AAAI Press,2007.540-545. 被引量:1
  • 5Liao XJ,Xue Y,Carin L.Logistic regression with an auxiliary data source.In:Proc.of the 22nd lnt*I Conf.on Machine Learning.San Francisco:Morgan Kaufmann Publishers,2005.505-512.[doi:10.1145/1102351.1102415]. 被引量:1
  • 6Xing DK,Dai WY,Xue GR,Yu Y.Bridged refinement for transfer learning.In:Proc.of the Ilth European Conf.on Practice of Knowledge Discovery in Databases.Berlin:Springer-Verlag,2007.324-335.[doi:10.1007/978-3-540-74976-9_31]. 被引量:1
  • 7Mahmud MMH.On universal transfer learning.In:Proc.of the 18th Int’l Conf.on Algorithmic Learning Theory.Sendai,2007.135-149.[doi:10,1007/978-3-540-75225-7_14]. 被引量:1
  • 8Samarth S,Sylvian R.Cross domain knowledge transfer using structured representations.In:Proc.of the 21st Conf.on Artificial Intelligence.AAAI Press,2006.506-511. 被引量:1
  • 9Bel N,Koster CHA,Villegas M.Cross-Lingual text categorization.In:Proc.of the European Conf.on Digital Libraries.Berlin:Springer-Verlag,2003.126-139.[doi:10.1007/978-3-540-45175-4_13]. 被引量:1
  • 10Zhai CX,Velivelli A,Yu B.A cross-collection mixture model for comparative text mining.In:Proc.of the 10th ACM SIGKDD Int’l Conf.on Knowledge Discovery and Data Mining.New York:ACM,2004.743-748.[doi:10.1145/1014052.1014150]. 被引量:1

共引文献137

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部 意见反馈