Emotion Classification in Textual Conversations Using Deep Broad Learning

In this chapter, we will explore emotion classification in textual conversations (ECTC). ECTC is one of the most important tasks for NLP in recent years. However, it is a challenging task for machines to conduct emotion classification in textual conversations, because emotions rely heavily on textual context. To address the challenge, we propose a method to classify emotion in textual conversations based on DL and BL. It aims to provide a better solution to capture the local contextual information (i.e., utterance-level) in an utterance, as well as the global contextual information (i.e., speaker-level) in a conversation by using CNN, Bi-LSTM, and BL. It is the first effort to combine DL and BL to classify emotion in textual conversations.