Significant advancements have recently been made in the fields of recommender systems and natural language processing, particularly with large language models (LLMs). In most cases, recommender systems have been used to suggest items and enhance personalization for users, while LLMs have been applied to textual tasks such as text completion, translation, and summarization. In this study, we demonstrate that integrating recommender system models with recent LLMs can effectively suggest appropriate surgical procedures for patients. We employ several LLMs to process clinical text in a morphologically rich language, serving three crucial roles: information representation, information enrichment, and explaining the surgical procedure suggestions made by the recommender system. Our method was evaluated using real-world clinical data, considering patients’ demographic attributes and health conditions. To assess the explainability of our method, we conducted an extensive experiment involving several clinicians. The results achieved by our method indicate that using recommender systems and LLMs can lead to high performance and improved explanations. Our study has the potential to enhance the personalization of healthcare and could be adopted by health services to assist healthcare professionals in recommending appropriate surgical procedures.

Modeling for information systems (IS) analysis and design offers broad insights into the advances and challenges of enterprise, business process, software, and conceptual modeling. In celebration of its 30th edition, this paper presents a data-driven retrospective analysis of studies published at the Exploring Modeling Methods for Systems Analysis and Development (EMMSAD) working conference from 2005 to 2024. EMMSAD has long been a key venue for research on Information Systems (IS) Modeling, covering areas such as conceptual modeling, enterprise modeling, and model-driven engineering, as well as the evaluation of modeling techniques and tools. Using machine learning, specifically Dynamic Topic Modeling (DTM) with BERTopic, this study identifies recurring topics, emerging trends, and shifts in research focus within the IS modeling community. The findings highlight key areas of alignment between IS modeling and the broader modeling landscape, providing insights into the field’s evolution and future research opportunities.