Abstract
Emotion detection-based applications are becoming popular in different areas, such as e-learning, mobile health-care and many other contexts. In this paper, we propose an emotion detection system from health-related content posted by users on mobile-based social media. The prior works in this area have focused on the formal emotion pointers, such as emotion words, valence shifters and negations, with limited coverage of informal emotion pointers, such as emoticons and slangs. The problems associated with the previous studies can be addressed more efficiently by classifying an emotion bearing text with the help of different classifiers in a sequential way. Our aim in this work is to develop a cognitive-based emotion detection system to classify emotions expressed by users (patients) in health reviews posted on mobile-enabled social media sites with emphasis on classifying emotion-based opinion words and informal emotion pointers, such as emoticons and slangs efficiently. The results obtained show the efficacy of the proposed system with respect to comparing methods. The system is versatile and can be extended to detect and classify emotions in other domains.