Dialog Act classification in Chinese spoken language

Dialog Act (DA) is an important pragmatics feature for us to understand speakers' intention. Many methods have been proposed to recognize DA tags. However, little work has been conducted to address the problem of DA tagging in Chinese spoken dialog language. In this work, we employ both the lexical features and the inter-utterance dependency features for DA tagging. And we propose three different methods: n-gram, extended HMM and n-gram+KNN. The experimental results show that these methods are effective for the task.