• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

计算机工程与科学

• 人工智能与数据挖掘 • 上一篇    下一篇

基于注意力机制的文本情感倾向性研究

裴颂文1,2,王露露1   

  1. (1.上海理工大学光电信息与计算机工程学院,上海 200093;2.复旦大学管理学院,上海 200433)
  • 收稿日期:2018-07-26 修回日期:2018-09-24 出版日期:2019-02-25 发布日期:2019-02-25
  • 基金资助:

    上海市浦江人才计划(16PJ1407600);中国博士后科学基金(2017M610230);国家自然科学基金(61332009,61775139);计算机体系结构国家重点实验室开放题目(CARCH201807)

Text sentiment analysis based on attention mechanism

PEI Songwen1,2,WANG Lulu1   


  1. (1.School of OpticalElectrical and Computer Engineering,University of Shanghai for Science and Technology,Shanghai 200093;
    2.School of Management,Fudan University,Shanghai 200433,China)
     
  • Received:2018-07-26 Revised:2018-09-24 Online:2019-02-25 Published:2019-02-25

摘要:

社交媒体上短文本情感倾向性分析作为情感分析的一个重要分支,受到越来越多研究人员的关注。为了改善短文本特定目标情感分类准确率,提出了词性注意力机制和LSTM相结合的网络模型PATLSTM。将文本和特定目标映射为一定阈值范围内的向量,同时用词性标注处理句子中的每个词,文本向量、词性标注向量和特定目标向量作为模型的输入。PATLSTM可以充分挖掘句子中的情感目标词和情感极性词之间的关系,不需要对句子进行句法分析,且不依赖情感词典等外部知识。在SemEval2014Task4数据集上的实验结果表明,在基于注意力机制的情感分类问题上,PATLSTM比其他模型具有更高的准确率。
 
 

关键词: 注意力机制, 长短时记忆网络, 短文本, 情感分析

Abstract:

As an important branch of sentiment analysis, short-text sentiment classification on social media has attracted more and more researchers’ attention. To improve the accuracy of the short text targetbased sentiment classification, we propose a network model that combines the part-of-speech attention mechanism with long short-term memory (PAT-LSTM). The text and the target are mapped to a vector  within a certain threshold range. In addition, each word in the sentence is marked by the part-of-speech. The text vector, target vector and part-of-speech vector are then input into the model. The PAT-LSTM model can fully explore the relationship between target words and emotional words in a sentence, and it does not require syntactic analysis of sentences or external knowledge such as sentiment lexicon. The results of comparative experiments on the Eval2014 Task4 dataset show that the PAT-LSTM network model has higher accuracy in attention-based sentiment classification.
 

Key words: attention mechanism, LSTM, short text, sentiment analysis