Toward Enhancing User Experience and Adoption of AI Chatbot Apps for Mental Health: Insights from Natural Language Processing of User Reviews

As mental health is essential for human well-being, the recent years have seen a proliferation in both the quantity and diversity of mobile health (mHealth) applications developed for mental healthcare. This paper presents an empirical study analyzing 144,165 user review comments from 20 mHealth apps within the mental health domain, collected from the Google Play Store. Utilizing frameworks for assessing mHealth technology adoption and employing natural language processing (NLP) techniques, the study aims to understand user perceptions and experiences from large sets of unstructured user reviews. The findings indicate an overall improvement in user sentiment toward mHealth apps over the years. Specifically, AI-powered mHealth apps (i.e., AI chatbots) received relatively more positive user reviews compared to non-AI apps, with users finding them to be more reliable, accessible, and user-friendly. However, issues related to intrusiveness, costs, and payment aspects still need improvement to meet users’ expectations. Finally, this study highlights the challenges and opportunities in designing human-centered AI (HCAI) systems to enhance user experiences and perceptions, particularly for at-risk populations.