A Taxonomy of Empathetic Response Intents in Human Social Conversations

Proceedings of the 28th International Conference on Computational Linguistics, pages 4886–4899 Barcelona, Spain (Online), December 8-13, 2020 4886 A Taxonomy of Empathetic Response Intents in Human Social Conversations Anuradha Welivita and Pearl Pu School of Computer and Communication Sciences ´Ecole polytechnique f´ed´erale de Lausanne Switzerland {kalpani.welivita,pearl.pu}@epfl.ch Abstract Open-domain conversational agents or chatbots are becoming increasingly popular in the natural language processing community. One of the challenges is enabling them to converse in an em- pathetic manner. Current neural response generation methods rely solely on end-to-end learning from large scale conversation data to generate dialogues. This approach can produce socially unacceptable responses due to the lack of large-scale quality data used to train the neural models. However, recent work has shown the promise of combining dialogue act/intent modelling and neural response generation. This hybrid method improves the response quality of chatbots and makes them more controllable and interpretable. A key element in dialog intent modelling is the development of a taxonomy. Inspired by this idea, we have manually labeled 500 response intents using a subset of a sizeable empathetic dialogue dataset (25K dialogues). Our goal is to produce a large-scale taxonomy for empathetic response intents. Furthermore, using lexical and machine learning methods, we automatically analysed both speaker and listener utterances of the entire dataset with identified response intents and 32 emotion categories. Finally, we use information visualization methods to summarize emotional dialogue exchange patterns and their temporal progression. These results reveal novel and important empathy patterns in human-human open- domain conversations and can serve as heuristics for hybrid approaches. 1 Introduction Inspired by the recent success of deep neural networks for natural language processing (NLP) tasks such as language modeling (Mikolov et al., 2010) and machine translation (Sutskever et al., 2014), neural response generation is currently at the forefront of research in the NLP community. Recent advances in this field have proven the efficacy of deep neural networks in modelling both task-oriented and open- domain dialogue systems (Wen et al., 2015; Sutskever et al., 2014; Vinyals and Le, 2015). Most of the existing neutral conversation models are capable of generating syntactically and contextually well- formed responses. Some of the work also focuses on enabling chatbots to generate emotionally colored and affect-rich responses (Asghar et al., 2018; Zhou et al., 2018; Xie et al., 2019). Despite the efforts in modeling affect in natural language, work that focuses specifically on modeling empathy in chatbots is relatively limited and remains an open research question (Spring et al., 2019). Empathy plays a vital role in human psychological processes for smooth social interaction (Decety, 2010). Empathy-related responding includes caring and sympathetic concerns for other people. Humans are born with core affect neural circuitry, and they gradually develop the ability to apprehend the emo- tional states of others and respond in an empathetic manner. Empathy motivates pro-social behavior and increases the sense of social bonding (Eisenberg and Eggum, 2009). Therefore, in the context of social interaction, a chatbot needs to be empathetic to maintain healthy interaction with …

Content truncated to 500 words