ANA at SemEval-2019 Task 3: Contextual Emotion detection in Conversations through hierarchical LSTMs and BERT

Proceedings of the 13th International Workshop on Semantic Evaluation (SemEval-2019), pages 49–53 Minneapolis, Minnesota, USA, June 6–7, 2019. ©2019 Association for Computational Linguistics 49 ANA at SemEval-2019 Task 3: Contextual Emotion detection in Conversations through hierarchical LSTMs and BERT Chenyang Huang, Amine Trabelsi, Osmar R. Za¨ıane Department of Computing Science, University of Alberta {chuang8,atrabels,zaiane}@ualberta.ca Abstract This paper describes the system submitted by ANA Team for the SemEval-2019 Task 3: EmoContext. We propose a novel Hierarchi- cal LSTMs for Contextual Emotion Detection (HRLCE) model. It classifies the emotion of an utterance given its conversational con- text. The results show that, in this task, our HRCLE outperforms the most recent state-of- the-art text classification framework: BERT. We combine the results generated by BERT and HRCLE to achieve an overall score of 0.7709 which ranked 5th on the final leader board of the competition among 165 Teams. 1 Introduction Social media has been a fertile environment for the expression of opinion and emotions via text. The manifestation of this expression differs from tradi- tional or conventional opinion communication in text (e.g., essays). It is usually short (e.g. Twit- ter), containing new forms of constructs, including emojis, hashtags or slang words, etc. This con- stitutes a new challenge for the NLP community. Most of the studies in the literature focused on the detection of sentiments (i.e. positive, negative or neutral) (Mohammad and Turney, 2013). Recently, emotion classification from social media text started receiving more attention (Yad- dolahi et al., 2017; Mohammad et al., 2018). Emo- tions have been extensively studied in psychology (Ekman, 1992; Plutchik, 2001). Their automatic detection may reveal important information in so- cial online environments, like online customer ser- vice. In such cases, a user is conversing with an automatic chatbot. Empowering the chatbot with the ability to detect the user’s emotion is a step forward towards the construction of an emotion- ally intelligence agent. Giving the detected emo- tion, an emotionally intelligent agent would gener- ate an empathetic response. Although its potential convenience, detecting emotion in textual conver- sation has seen limited attention so far. One of the main challenges is that one users utterance may be insufficient to recognize the emotion (Huang et al., 2018). The need to consider the context of the conversion is essential in this case, even for human, specifically given the lack of voice mod- ulation and facial expressions. The usage of figu- rative language, like sarcasm, and the class size’s imbalance adds up to this problematic (Chatterjee et al., 2019a). Context LSTM Context LSTM Context LSTM Utterance Encoder Multi-head self-attention Category Classification Utterance LSTM GloVe ELMo DeepMoji Utterance Encoder Utterance Encoder Utterance Encoder Figure 1: An illustration of the HRLCE model In this paper, we describe our model, which was proposed for the SemEval 2019-Task 3 competi- tion: Contextual Emotion Detection in Text (Emo- Context). The competition consists in classify- ing the emotion of an utterance given its conver- sational context. More formally, given a textual user utterance along with 2 turns of context …

Content truncated to 500 words