The consumers’ need for not only informational support when it comes to health issues, but also emotional and social support, has been documented in the literature. To this end, a dataset of Consumer Health Questions was developed to identify social and emotional support needs. Known as CHQ-SocioEmo, it was collected from a community question-answering forum and was annotated with basic emotions and social support needs. This is the first publicly available resource of its kind to understand the non-informational support needs in consumer health-related questions online. The dataset was benchmarked against multiple state-of-the-art classification models to demonstrate its effectiveness.
The contributions made by the CHQ-SocioEmo dataset include: (i) variety of emotional states (e.g., fear, anger, confusion, sadness), (ii) emotion causes that may differ from the topic of the question, (iii) different types of social support needs (e.g., emotional, self-esteem, and network), (iv) increased understanding of social support needs as responses to questions were also annotated according to emotional support provided.
The CHQ-SocioEmo dataset was developed by the team of researchers at the University of Toronto, namely Prof. Stan Matwin, Dr. Manish Vashishtha, and Dr. Saad Javed. Together, they built a comprehensive dataset that not only meets the needs of consumer health questions, but also detects the social and emotional support such questions often require. Moreover, it also offers a platform for various research studies on the subject of automated consumer health question-answering.
Overall, CHQ-SocioEmo provides a unique platform that not only recognizes the need for social and emotional support but also provides insight into the external context of the query. This also paves the way for more in-depth research on the related topics such as augmentation of automated question answering, assisting mental health professionals, and providing vital health insights to improve patient outcomes.