Attention Model Using Full-Time Information For Sentences Emotion Classification
Chiyu Wang, Hong Li, Xun Hu, Jiale Zhou · 2019
This paper proposed a method that uses the output information of LSTM at each time point for the final sentence emotional judgment, and constructed three classification models based on the existing attention model and different filters. In this paper, the model was trained on the basis of pre-trained word vectors and compared with the existing models. Through rigorous experiments, it is found that the performance of the model improved by this article is better than the model that does not use full-time information in emotional classification, and models using different filters show different results.