Chatting activity recognition in social occasions using factorial conditional random fields with iterative classification

摘要: Recognizing activities in social occasions plays an important role of building human networks. For example, the recognition interactions could be great help to determine whether any two attendees have same interests academic conference or a cocktail party. Among various types interactions, chatting with others is significant indicator. Furthermore, duration activity may imply strength interaction reality. It therefore recognize patterns occasions. During real-world conversation, person often begins talking following other person’s utterance completed. Linguistic experts observed that usually performed as interlaced dialogic process. As result, it intuitive apply dynamic probabilistic models learning and detecting activities.