Silva-Aravena, FabiánFabiánSilva-AravenaMorales, JennyJennyMoralesSáez, PaulaPaulaSáezCornide-Reyes, HéctorHéctorCornide-ReyesBaltierra, SergioSergioBaltierra2026-10-082026-10-082026Silva-Aravena, Fabián; Morales, Jenny; Sáez, Paula; Cornide-Reyes, Héctor; Baltierra, Sergio (2026). Unsupervised Clustering to Segment Patient Experience Under Waiting and Access Constraints. Lecture Notes in Computer Science, 16716, 469-483. https://doi.org/10.1007/978-3-032-31418-5_290302-97431611-3349https://hdl.handle.net/20.500.12740/24916Patient experience is a relevant indicator for evaluating the quality of healthcare, reflecting patients perceptions of the entire care process. In various hospitals, the patient experience is also influenced by organizational factors such as waiting times and access delays. Therefore, in this study, we propose an unsupervised clustering approach to segment patient experience using data from a survey of 1,000 users of a public hospital in Pakistan. Dimensions related to experience include registration and reception, medical consultation, laboratory and pharmacy services, and healthcare infrastructure. These dimensions indirectly capture the cumulative effect of waiting and service flow along the patient care trajectory. Before clustering, missing values were imputed using median statistics, and ordinal variables were normalized using z-scores. We used the K-means algorithm, which identified two groups of patients. The optimal solution was determined using the Elbow and Silhouette criteria: a low-to medium experience group and a high-experience group. Patients in the high-experience group received high ratings for staff courtesy, pharmacy efficiency, and environmental cleanliness. For post-hoc validation, we used the Overall Satisfaction Index (OSI), which showed that 91.9% of patients in the highly experienced group reported high levels of satisfaction. This result demonstrates significant agreement with the methodology for supporting clinical decision-making.http://purl.org/coar/access_right/c_14cbPatient ExperienceUnsupervised LearningK-Means ClusteringHealthcare QualityService SegmentationBook Chapterhttps://doi.org/10.1007/978-3-032-31418-5_29