Sentiment classification using attention mechanism and bidirectional long short-term memory network

We propose a sentiment classification method for large scale microblog text based on the attention mechanism and the bidirectional long short-term memory network (SC-ABiLSTM). We use an experimental study to compare our proposed method with baseline methods using real world large-scale microblog data. Comparing the accuracy of the baseline methods to the accuracy of our model, we demonstrate the efficacy of our proposed method. While sentiment classification of social media data has been extensively studied, the main novelty of our study is the implementation of the attention mechanism in a deep learning network for analyzing large scale social media data. •Sentiment classification based on Attention-based Bidirectional long short-term memory network.•A mix embedding model for sentiment classification based on the Word2Vec and GloVe.•Propose an effective sentiment classification method for large-scale microblog text.