Book excerpt: As time goes on, big companies such as Amazon, Microsoft, Google and Apple become increasingly interested in virtual assistants. The interest and development of social robots has put research into affective and social computing at the forefront of the scene. The aim of Opinion Analysis in Interactions is to present methods based on artificial intelligence through a combination of machine learning models and symbolic approaches. Also discussed are natural language processing and affective computing, via the analysis and generation of socio-emotional signals. The book explores the analysis of opinions in human–human interaction and tackles the less-explored (yet crucial) challenges related to the analysis methods of user opinions within the context of human–agent interaction. It also illustrates the implementation of strategies for selecting and generating agent utterances in response to user opinions, and opens up perspectives on the agent’s multimodal generation of utterances that hold attitudes.