Mapping the research landscape of recommender systems for digital libraries

A bibliometric analysis of two decades (2004-2023)

Recommender System Digital Library Information Overload Fuzzy Linguistic Modeling Deep Learning Collaborative Filtering

Authors

June 22, 2024

Downloads

Background of the study: In today's information-rich environment, researchers face the difficulty of managing information overload and struggle to identify meaningful information among a plethora of choices. A recommender system plays a crucial role in assisting users in discovering relevant information from a digital library.

Purpose: This study aims to provide a bibliometric analysis of research publications on recommender systems for digital libraries.

Method: The research method involved the quantitative bibliometric approach to analyse the research publications from the Web of Science and Scopus databases using Biblioshiny. The dataset retrieved comprises 374 documents published between the period of 2004-2023.

Findings: The study's findings highlight that the number of publications was notably high in 2009 and 2018 within the analyzed period. The most significant contributors are Porcel C, Herrera-Viedma E, and Beel J. There is a considerable international collaboration between the countries- China, USA, and Germany. 

Conclusions: This study indicates an emerging interest in recommender systems for digital libraries with the continuous evolution of new recommendation models. There is a huge potential for research considering the availability of multimodal data, the continuous evolution of new technologies such as deep learning, and opportunities for the development of an architecture for integrating different digital libraries.