Meta-learning-based lightweight learning framework for healthcare recommendation system

Abstract

The recommendation system provided effective results in various industries. Despite the notable results, the recommendation system in healthcare had significant difficulties in collecting, utilizing, and analyzing data due to privacy laws. As improvements in the Data 3 Act have expanded the scope of data utilization, the need for a framework that is easy to leverage and learn has increased. Our model collects data in a variety of IOT environments and utilizes meta-learning to solve cold start problems and provides a way to quickly converge on various recommended system models.