Hierarchical state recurrent neural network for social emotion ranking

Text generation with auxiliary attributes, such as topics or sentiments, has made remarkable progress. However, high-quality labeled data is difficult to obtain for the large-scale corpus. Therefore, this paper focuses on social emotion ranking aiming identify emotions different intensities evoked by online documents, which could be potentially beneficial further controlled text generation. Existing studies often consider each document an entirety that fail capture inner relationship between sentences in a document. In paper, we propose novel hierarchical state recurrent neural network ranking. A hierarchy mechanism employed key semantic structure Moreover, instead of incrementally reading sequence words traditional networks, proposed approach encodes hidden states all simultaneously at step long-range dependencies precisely. Experimental results show performs remarkably better than state-of-the-art approaches and useful