EmotionX-KU: BERT-Max based Contextual Emotion Classifier
This work addresses emotion analysis in conversational AI, offering incremental improvements for dialogue systems.
The authors tackled emotion classification in dialogues by proposing a BERT-Max model with dynamic max pooling and weighted loss, achieving state-of-the-art results on the Friends and EmotionPush datasets and competitive performance in the EmotionX 2019 challenge.
We propose a contextual emotion classifier based on a transferable language model and dynamic max pooling, which predicts the emotion of each utterance in a dialogue. A representative emotion analysis task, EmotionX, requires to consider contextual information from colloquial dialogues and to deal with a class imbalance problem. To alleviate these problems, our model leverages the self-attention based transferable language model and the weighted cross entropy loss. Furthermore, we apply post-training and fine-tuning mechanisms to enhance the domain adaptability of our model and utilize several machine learning techniques to improve its performance. We conduct experiments on two emotion-labeled datasets named Friends and EmotionPush. As a result, our model outperforms the previous state-of-the-art model and also shows competitive performance in the EmotionX 2019 challenge. The code will be available in the Github page.