MoonGrad at SemEval-2019 Task 3: Ensemble BiRNNs for Contextual Emotion Detection in Dialogues

Proceedings of the 13th International Workshop on Semantic Evaluation (SemEval-2019), pages 261–265 Minneapolis, Minnesota, USA, June 6–7, 2019. ©2019 Association for Computational Linguistics 261 MoonGrad at SemEval-2019 Task 3: Ensemble BiRNNs for Contextual Emotion Detection in Dialogues Chandrakant Bothe and Stefan Wermter Knowledge Technology, Department of Informatics, University of Hamburg, Vogt-Koelln-Str. 30, 22527 Hamburg, Germany www.informatik.uni-hamburg.de/WTM/ {bothe, wermter}@informatik.uni-hamburg.de Abstract When reading “I don’t want to talk to you any more”, we might interpret this as either an an- gry or a sad emotion in the absence of context. Often, the utterances are shorter, and given a short utterance like “Me too!”, it is difficult to interpret the emotion without context. The lack of prosodic or visual information makes it a challenging problem to detect such emo- tions only with text. However, using contex- tual information in the dialogue is gaining im- portance to provide a context-aware recogni- tion of linguistic features such as emotion, di- alogue act, sentiment etc. The SemEval 2019 Task 3 EmoContext competition provides a dataset of three-turn dialogues labeled with the three emotion classes, i.e. Happy, Sad and An- gry, and in addition with Others as none of the aforementioned emotion classes. We develop an ensemble of the recurrent neural model with character- and word-level features as an in- put to solve this problem. The system per- forms quite well, achieving a microaveraged F1 score (F1μ) of 0.7212 for the three emotion classes. 1 Introduction Humans might interpret text wrongly when read- ing sentences in the absence of context, so ma- chines might too. When reading the following ut- terance, Why don’t you ever text me? it is hard to interpret the emotion where it can be either a sad or an angry emotion (Chatterjee et al., 2019; Gupta et al., 2017). The problem becomes even harder when there are ambiguous utterances, for example, the following utterance: Me too! one cannot really interpret the emotion behind such an utterance in the absence of context. See Table 1 where the utterance “Me too!” is used in many emotional contexts such as sad, angry, and happy and also in the class “others” where none of aforementioned emotions is present. Analyzing the emotion or sentiment of text pro- vides the opinion cues expressed by the user. Such cues could assist computers to make better deci- sions to help users (Kang and Park, 2014) or to prevent potentially dangerous situations (O’Dea et al., 2015; Mohammad and Bravo-Marquez, 2017; Sailunaz et al., 2018). Character-level deep neural networks have recently showed outstand- ing results on text understanding tasks such as machine translation and text classification (Zhang et al., 2015; Kalchbrenner and Blunsom, 2013). Usually, the utterances are short and contain mis-spelt words, emoticons, and hashtags, espe- cially in the textual conversation. Hence, using character-level language representations can the- oretically capture the notion of such texts. On the other hand, the EmoContext dataset is col- lected from the social media, and so the charac- ter language model used in our experiments is also …

Content truncated to 500 words