Evolution of Scientific Interests: A Time-Aware Recommender System

The increasing volume of scientific publications makes it challenging for researchers to stay updated. Traditional recommendation systems often overlook the evolving nature of academic interests, leading to outdated suggestions. This article introduces a time-aware recommendation system (TARS) that incorporates temporal dynamics to improve the relevance of scientific article recommendations. In experiments with OpenAlex data, TARS outperformed static models, achieving up to 12% better precision and recall, particularly for active researchers. The study highlights the importance of temporal modeling in capturing evolving preferences and enhancing engagement in academic information systems.