Emotion analytics: Design and evaluation of context-aware affective systems

Understanding human emotions is fundamental to our daily interactions. While emotions are quite intuitively understood by humans, it is rather tricky to train machines to understand emotions under a wide variety of naturalistic conditions. My collaborators and I work on building and validating conceptual and computational models of emotion perception and recognition. As part of our research in socially intelligent systems at IHPC, we have developed capabilities to perform fine-grained emotion detection from various kinds of textual, vocal as well as visual expressions. Such large scale emotion data is extremely valuable in a variety of social and business contexts *(e.g. design of empathetic chatbots, detecting misinformation and online hate, predicting popularity of online video ads, performing quality assessment of customer service interactions etc.)*

Recent recognition: "Digital Emotions" at IHPC (A*STAR), was a finalist at IET Innovation Awards 2018, "Intelligent Systems" category (IETs annual Innovation Awards recognize the best in new innovations in science, engineering and technology)

Selected projects/papers:
1. Exploring the contextual factors affecting multimodal emotion recognition in videos, with Raj Kumar Gupta, and Yang Yinping (Published in IEEE Transactions on Affective Computing)
2. What constitutes happiness? predicting and characterizing the ingredients of happiness using emotion intensity analysis, with Raj Kumar Gupta, and Yang Yinping. (Presented at AffCon@ AAAI)