Abstract

Since CVDs are the leading cause of death worldwide, this article solves the problem of developing a recommendation algorithm for the prevention of cardiovascular diseases (CVDs), as the main component of health recommender systems. To address this issue, a knowledge-based recommendation algorithm was proposed to support self-management of CVD risk factors in adults at home. The proposed algorithm is based on a new user profile and an original multidimensional recommendation model. The user profile includes not only descriptive but also predictive assessments of cardiovascular health based on the SCORE model which are outlined in official guidelines. A multidimensional recommendation model contains target, informational, and explanatory components for higher health lifestyle adherence. We found that ChatGPT provided plausible and understandable explanations of cardiovascular disease risk factors. therefore, these explanations were used in the texts of the recommendations. The main feature of the proposed algorithm is the combination of rule-based logic with the capabilities of a large language model to generate human-like text of multidimensional recommendations. Evaluation of the proposed algorithm showed high user satisfaction with the proposed recommendation algorithm.

Statistics