A Multi-Sentiment Classifier Based on GRU and Attention Mechanism

Previous sentiment analysis studies have focused on monolingual texts, and are basically multi-category tasks (ie, a sentence belongs to only one category). However, in practice, a sentence often expresses multiple sentiments, and the text often contains multiple languages. This paper proposes a multi-label sentiment classifier based on GRU and attention mechanism, which has achieved good results in the data set provided by NLP&CC share task 1.