A complex activity based emotion recognition algorithm for affect aware systems

The usefulness of affect aware systems can be summarized as optimizing the system's services to improve the user experience. A means to achieve this objective is to make the system aware of the real-world situations and enable intelligent emotion analysis of its users. The ability of smart systems to automatically recognize human affective behavior through their movements and to respond are the key challenges in this area of study. This paper thus proposes an algorithm that can effectively categorize the underlining emotion of an activity as positive or negative and then classify it amongst the six basic emotions - sadness, fear, happiness, anger, disgust and surprise [1]. To portray the efficacy of this proposed model, typical scenes in the context of a smart home have been analyzed and the results have been presented. The results presented show that the proposed model would help developers to understand user experiences in the form of user's emotions in any scenario, which would help them to improve or modify the existing technologies to enhance computerhuman interaction.