Abstract :
Sentiment analysis (SA) or opinion mining is the computational study of opinions, sentiments, attitudes, emotions, moods and affects. Due to almost unlimited applications and numerous research challenges, SA is a very active research area in natural language processing and data mining. In this talk, I will first introduce the SA problem and discuss the recent work of using big data and lifelong learning to help solve the problem. I then discuss a consumer application of sentiment analysis. This naturally leads to the interesting topic of intelligent personal assistants such as Siri, Cortana, and Google Now, and Chatbots such as XiaoIce. I believe such systems will be the next big thing of AI and will profoundly change our lives. For these systems to be very useful and widely adopted, they need to perform extensive opinion, sentiment, and emotion analysis, and to converse with humans affectively.
