본문 바로가기
  • Home

A Systematic Review of Deep Learning-Based Article Recommendation Research

  • Journal of the Korean Society for Library and Information Science
  • 2025, 59(1), pp.461-488
  • DOI : 10.4275/KSLIS.2025.59.1.461
  • Publisher : 한국문헌정보학회
  • Research Area : Interdisciplinary Studies > Library and Information Science
  • Received : January 28, 2025
  • Accepted : February 13, 2025
  • Published : February 28, 2025

Kim Seonghun 1

1성균관대학교

Excellent Accredited

ABSTRACT

Finding relevant research papers for citation has become increasingly challenging due to information overload, leading to growing interest in research paper recommendation systems. Over the past decade, deep learning methods have been employed to enhance the quality of paper recommendations. However, there has been a lack of systematic research on deep learning-based paper recommendation systems since 2020. This study conducts a systematic literature review of 47 deep learning-based research paper recommendation studies published between 2020 and 2023. The analysis examines data factors, data representation methods, methodologies, recommendation types, encountered challenges, and personalization aspects. Furthermore, it compares these studies with pre-2020 research by summarizing recommendation techniques, utilized datasets, deep learning methods, and key research topics by year. Based on this comparison, key insights are derived. Additionally, future research directions are suggested based on the analyzed research trends.

Citation status

* References for papers published after 2023 are currently being built.

This paper was written with support from the National Research Foundation of Korea.