CHQ- SocioEmo: Identifying Social and Emotional Support Needs in Consumer-Health Questions

Area of Science:

• Health Informatics
• Natural Language Processing
• Social Computing

Background:

• The public increasingly seeks health information online, often requiring more than just factual answers.
• Existing research and datasets focus on informational needs in consumer health questions, neglecting social and emotional support.
• Automated systems for consumer health questions must recognize and address non-informational needs.

Purpose of the Study:

• To introduce CHQ-SocioEmo, the first publicly available annotated dataset for non-informational needs in consumer health questions.
• To provide a resource for understanding and classifying social and emotional support requirements in online health queries.
• To benchmark the dataset's utility with state-of-the-art classification models.

Main Methods:

• Collected consumer health questions from a community question answering forum.
• Annotated the collected questions with basic emotions and social support needs.
• Developed and benchmarked the CHQ-SocioEmo dataset against various classification models.

Main Results:

• The CHQ-SocioEmo dataset is the first resource of its kind for non-informational needs in consumer health.
• Benchmarking demonstrated the dataset's effectiveness in training and evaluating classification models for social and emotional support needs.
• The dataset facilitates advancements in automated consumer health question answering systems.

Conclusions:

• Addressing social and emotional support is crucial for satisfactory online health information seeking.
• The CHQ-SocioEmo dataset provides a valuable resource for developing more comprehensive consumer health question-answering systems.
• Future research can leverage this dataset to improve AI's ability to provide empathetic and supportive responses to health queries.