Year: 2025
Type: article
Abstract: With the swift progress of intelligent health management, health recommendation systems have become a crucial part of the healthcare area. Yet, traditional recommendation systems have many issues, such as data sparsity, inaccurate feature extraction, and insufficient personalization. This paper presents an optimized health recommendation algorithm based on the Transformer model, aiming to enhance the accuracy and user experience of health recommender systems. It combines the strengths of Graph Neural Networks (GNNs) and the Transformer model. By aggregating users' historical behavior data, it can capture potential features in health information and improve recommendations. Experimental results show that the proposed algorithm outperforms traditional models in recommendation accuracy and user satisfaction.
Source: Biomedical Data Science
Author: Xiaolong Luo
Institution: Southwest University
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4413026439