Examining user-AI interaction patterns in health-Information queries

Area of Science:

• Health Informatics
• Artificial Intelligence in Healthcare
• Human-Computer Interaction

Background:

• Growing use of generative artificial intelligence (GAI) tools for information retrieval.
• Limited understanding of how individuals engage with GAI for health-related queries.
• Need to analyze real-world user interactions with GAI in the health domain.

Purpose of the Study:

• To investigate patterns of user interaction with GAI for health information seeking.
• To categorize the types of health-related questions users pose to GAI.
• To understand the role of GAI as a health information resource and support tool.

Main Methods:

• Analysis of a dataset comprising user-GAI chat logs.
• Application of data and text-analytic methods for interaction categorization.
• Utilization of supervised machine learning, specifically Support Vector Machines (SVMs), for theme identification.

Main Results:

• Users frequently employ GAI for symptom exploration and medical education.
• GAI is utilized for mental health support and general health advice.
• Identified recurrent themes in user queries and interactions with GAI.

Conclusions:

• GAI tools serve as significant informational resources for health queries.
• GAI can function as a preliminary support tool, influencing user health knowledge.
• GAI interactions may encourage users to seek professional medical consultations.