A paper co-authored by Mr. Kavindi—a student at The Kyoto College of Graduate Studies for Informatics (KCGI), part of UOI - THE UNIVERSITY OF INFORMATICS, and originally from Sri Lanka—along with his advisors, Associate Professor Dinh Duy Tai and Professor Wuyi Yue, was published on September 23, 2026–25, 2026, in Paphos, Cyprus, at the 15th International Conference on Model and Data Engineering (MEDI 2026).The paper is scheduled to be published in *Lecture Notes in Computer Science* (LNCS), a renowned book series in the field of computer science published by Springer in Germany.
This paper summarizes the results of a study that analyzed online reviews of pediatric medical care using a combination of AI-based sentiment analysis and qualitative analysis, thereby visualizing characteristics of the “patient experience” that are difficult to capture through star ratings alone.We conducted sentiment analysis using five AI models and qualitative analysis using the MAXQDA qualitative data analysis software on 4,086 Google Maps reviews posted for 110 pediatric medical facilities in Kyoto City.
As a result, while a strong correlation was found between star ratings and the sentiments inferred from the text of the reviews, it became clear that, particularly in 3-star reviews, positive and negative experiences coexist in a complex manner.Furthermore, in positive reviews, the kindness and expertise of medical staff, the sense of security they provide, trust in their care, and ease of access to care were highlighted as key factors. In negative reviews, the main complaints centered on interpersonal interactions, wait times, accessibility, and concerns regarding the quality and safety of care.
This study demonstrated the potential to use online reviews not merely as a “star rating,” but as a source of information for understanding how patients and their families experience pediatric care, what they value, and what challenges they perceive.By combining large-scale text analysis using AI with qualitative analysis, we have proposed a new evaluation and analysis method that can lead to improvements in pediatric healthcare services.