BiERU: Bidirectional emotional recurrent unit for conversational sentiment analysis

• We propose a fast, compact and parameter-efficient party-ignorant framework based on emotional recurrent unit. design generalized neural tensor block which is suitable for different structures, to perform context compositionality. Experiments three standard benchmarks indicate that our model outperforms the state of art with fewer parameters. Sentiment analysis in conversations has gained increasing attention recent years growing amount applications it can serve, e.g., sentiment analysis, recommender systems, human-robot interaction. The main difference between conversational single sentence existence information may influence an utterance dialogue. How effectively encode contextual dialogues, however, remains challenge. Existing approaches employ complicated deep learning structures distinguish parties conversation then information. In this paper, we named bidirectional unit analysis. system, followed by two-channel classifier designed compositionality classification, respectively. Extensive experiments datasets demonstrate most cases.