新闻和用户生成内容的交互分析.pptVIP

  • 28
  • 0
  • 约2.64千字
  • 约 22页
  • 2017-06-15 发布于天津
  • 举报
新闻和用户生成内容的交互分析.ppt

Linguistically Regularized LSTM for Sentiment Classification Qiao Qian, Minlie Huang, Jinhao Lei, Xiaoyan Zhu Tsinghua University ACL 2017 Outline Introduction Model Experiment Conclusion Background Sentiment Classification: positive or negative more fine-grained classes Examples: The movie is very interesting. Strong Positive I saw this movie yesterday. Neutral Its a little boring. Weakly Negative Related Work Neural Networks Recurrent Neural Network / LSTM [Hochreiter and Schmidhuber, 1997] Convolutional Neural Network [Kim, 2014; Kalchbrenner et a

文档评论(0)

1亿VIP精品文档

相关文档