Discovering Emotion and Reasoning its Flip in Multi-Party Conversations using Masked Memory Network and Transformer

Efficient discovery of a speaker's emotional states in a multi-party\nconversation is significant to design human-like conversational agents. During\na conversation, the cognitive state of a speaker often alters due to certain\npast utterances, which may lead to a flip in their emotional state. Therefore,\ndiscovering the reasons (triggers) behind the speaker's emotion-flip during a\nconversation is essential to explain the emotion labels of individual\nutterances. In this paper, along with addressing the task of emotion\nrecognition in conversations (ERC), we introduce a novel task - Emotion-Flip\nReasoning (EFR), that aims to identify past utterances which have triggered\none's emotional state to flip at a certain time. We propose a masked memory\nnetwork to address the former and a Transformer-based network for the latter\ntask. To this end, we consider MELD, a benchmark emotion recognition dataset in\nmulti-party conversations for the task of ERC, and augment it with new\nground-truth labels for EFR. An extensive comparison with five state-of-the-art\nmodels suggests improved performances of our models for both tasks. We further\npresent anecdotal evidence and both qualitative and quantitative error analyses\nto support the superiority of our models compared to the baselines.\n