DualSentiNet: Dual prediction of word and document sentiments using shared word embedding

With the popularization of social networking services, numerous words are newly emerging every day in personalized document sources. Slang terms, abbreviations, newly coined words, and nongrammatical words or expressions belong here, and people are more likely to use these words with a certain sentimental tendency compared to other standard words. Thus, it becomes important to nd their meanings or sentiments to analyze the sentiment of user-generated texts. This paper proposes a novel sentiment analysis model, termed DualSentiNet, which predicts the sentiments of newly emerged words and documents at the same time. Our model is composed of three parts: (i) a word-level sentiment regression network, (ii) a document-level sentiment classi cation network, and (iii) a shared word embedding layer. DualSentiNet makes a word embedding layer shared by two different networks, thereby learning richer information about both word-level and document-level sentiments through two-way back-propagation. Consequently, it improves the performance of sentiment prediction by preventing word vectors from being over tted. Experimental results show that DualSentiNet signi cantly outperforms competitors in terms of both document sentiment classi cation accuracy and the word sentiment regression RMSE. In addition, DualSentiNet produces better word embedding by re"ecting both word and document sentiments.