ntuer at SemEval-2019 Task 3: Emotion Classification with Word and Sentence Representations in RCNN

Proceedings of the 13th International Workshop on Semantic Evaluation (SemEval-2019), pages 282–286 Minneapolis, Minnesota, USA, June 6–7, 2019. ©2019 Association for Computational Linguistics 282 ntuer at SemEval-2019 Task 3: Emotion Classification with Word and Sentence Representations in RCNN Peixiang Zhong, Chunyan Miao Joint NTU-UBC Research Centre of Excellence in Active Living for the Elderly Nanyang Technological University Singapore peixiang001@e.ntu.edu.sg, ascymiao@ntu.edu.sg Abstract In this paper we present our model on the task of emotion detection in textual conver- sations in SemEval-2019. Our model ex- tends the Recurrent Convolutional Neural Net- work (RCNN) by using external fine-tuned word representations and DeepMoji sentence representations. We also explored several other competitive pre-trained word and sen- tence representations including ELMo, BERT and InferSent but found inferior performance. In addition, we conducted extensive sensitiv- ity analysis, which empirically shows that our model is relatively robust to hyper-parameters. Our model requires no handcrafted features or emotion lexicons but achieved good perfor- mance with a micro-F1 score of 0.7463. 1 Introduction Emotions are psychological and physiological states generated in humans in reaction to inter- nal or external events. Messages in human con- versations inherently convey emotions. With the rise of social media platforms such as Twitter, as well as chatbots such as Amazon Alexa, there is an emerging need for machines to understand human emotions in conversations, which has a wide range of applications such as opinion analysis in cus- tomer support (Devillers et al., 2002) and provid- ing emotion-aware responses (Zhong et al., 2019). SemEval-2019 Task 3: EmoContext (Chatterjee et al., 2019b) is designed to promote research in this task. This task is to detect emotions in textual conver- sations. Each conversation is composed of three turns of utterances and the objective is to detect the emotion of the last utterance given the first two utterances as the context. The emotions in this classification task include happy, sad, angry and others, adapted from the well-known Ekman’s six basic emotions: anger, disgust, fear, happiness, sadness, and surprise (Ekman, 1992). The evalu- ation criteria is micro-averaged F1 score since the data is extremely unbalanced, as shown in Table 1. In recent years, pre-trained word and sentence representations achieved very competitive per- formance in many NLP tasks, e.g., fine-tuned word embeddings using distant training (Cliche, 2017) and tweet sentence representations Deep- Moji (Felbo et al., 2017) on sentiment analysis, and contextualized word representations BERT (Devlin et al., 2018) on 11 NLP tasks. Motivated by these successes, in this task we explored differ- ent word and sentence representations. We then fed these representations into a Recurrent Con- volutional Neural Network (RCNN) (Lai et al., 2015) for classification. RCNN includes a Long short-term memory (LSTM) network (Hochreiter and Schmidhuber, 1997) to capture word order- ing information and a max-pooling layer (Scherer et al., 2010) to learn discriminative features. We also experimented LSTM and CNN in our pre- liminary analysis but achieved worse performance as compared to RCNN. Our final system adopted fine-tuned word embeddings and DeepMoji as our choices of word and …

Content truncated to 500 words