Utterance-level Emotion Recognition in Conversation with Conversation-level Supervision

Emotion Recognition in Conversations (ERC) involves automatically identifying the emotion of each utterance in conversations. The emotion of an utterance is contingent to the conversation context, and thus, annotating each utterance in ERC entails repetitive screening the whole conversation from annotators. Such a requirement leads to prohibitive cost in fine-grained labeling on utterance. In this paper, we propose an efficient coarse-grained labeling strategy for ERC, which assigns a set of emotions for each conversation. In specific, we reformulate the ERC predictors with conversation-level emotion sets as weakly-supervised learning to optimise a potential candidate for ERC, which is termed as Dataless ERC (DERC). To validate this, we propose a simple-yet-flexible DERC framework with Progressive Learning (DERC-PL). We jointly update pseudo-utterance-level emotions and the ERC predictor in a self-training manner, where we progressively update the ERC predictor from training subsets with lower noise densities to the ones with higher noise densities. We implemented several versions of DERC-PL by incorporating various off-the-shelf ERC methods. Extensive experimental results demonstrate that the proposed DERC-PL can be on par with existing weakly-supervised learning baselines and supervised learning ERC methods.