短文本情感分析，在舆情监控和商业上有很多重要应用。以微博评论文本为研究对象，通过对微博评论文本进行分词、去除停用词，并使用Word2vec进行词向量训练得到词向量，并在Bi-LSTM中引入Attention机制，对Bi-LSTM双向处理后的结果进行加权进行输出。实验结果表明，Attention-Bi-LSTM与Bi-LSTM相比能有效识别出情感语句中重要的语义，提高预测的准确度。
Sentiment analysis has many important applications in public opinion monitoring and business. In this paper, microblog comment text is taken as the research object. Word segmentation is carried out on microblog comment text, stopping words are removed, and Word2vec is used for word vector training to obtain the word vector. Attention mechanism is introduced in Bi-LSTM, and the results of Bi-LSTM two-way processing are weighted and output. Experiments show that Attention-Bi-LSTM can effectively identify the important semantics of emotional statements and improve the accuracy of prediction compared with Bi-LSTM.
Pal, S., Ghosh, S. and Nag, A. (2018) Sentiment Analysis in the Light of LSTM Recurrent Neural Net-works. International Journal of Synthetic Emotions, 9, 33-39. https://doi.org/10.4018/IJSE.2018010103
[3]
Lu-ong, M.T., Pham, H. and Manning, C.D. (2015) Effective Approaches to Attention-Based Neural Machine Translation. Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, Lisbon, September 2015, 1412-1421. https://doi.org/10.18653/v1/D15-1166
[4]
Zhou, P., Shi, W., Tian, J., et al. (2016) Attention-Based Bidirectional Long Short-Term Memory Networks for Relation, Classification. Meeting of the Association for Computa-tional Linguistics. https://doi.org/10.18653/v1/P16-2034
