Emily: Developing An Emotion-affective Open-Domain Chatbot with Knowledge Graph-based Persona
This addresses the challenge of making chatbots more empathetic and consistent for users in open-domain conversations, though it appears incremental by building on existing dialogue models and knowledge graph techniques.
The paper tackles the problem of creating an emotion-aware open-domain chatbot that can detect negative user emotions and provide support by positively influencing emotional states, while maintaining personality consistency through knowledge graph-based handling of personal information. It shows that Emily outperforms state-of-the-art chatbots in emotion affecting and addressing personality inconsistency.
In this paper, we describe approaches for developing Emily, an emotion-affective open-domain chatbot. Emily can perceive a user's negative emotion state and offer supports by positively converting the user's emotion states. This is done by finetuning a pretrained dialogue model upon data capturing dialogue contexts and desirable emotion states transition across turns. Emily can differentiate a general open-domain dialogue utterance with questions relating to personal information. By leveraging a question-answering approach based on knowledge graphs to handle personal information, Emily maintains personality consistency. We evaluate Emily against a few state-of-the-art open-domain chatbots and show the effects of the proposed approaches in emotion affecting and addressing personality inconsistency.