<![CDATA[Objective: This study maps trends, approaches, challenges, and future research directions in digital library recommendation systems. Theoretical framework: The study focuses on recommendation systems in digital libraries, exploring Collaborative Filtering (CF), Content-Based Filtering (CBF), and hybrid approaches. It emphasizes algorithm optimization to address data sparsity and cold-start issues, and the integration of deep learning for improved accuracy and personalization. Literature review: The literature review tracks the evolution of recommendation systems from CF and CBF to hybrid and deep learning models, focusing on accuracy and cold-start issues. It highlights the growing use of advanced models and the challenges of algorithm optimization and data scarcity. Methods: A Systematic Literature Review (SLR) was conducted following the PRISMA framework. Literature was searched on Scopus using keywords related to recommendation systems. Data was analyzed using RStudio with Bibliometrix and VOSviewer for keyword network visualization. Results: This study shows a significant trend in the development of digital library recommendation systems, with publications increasing rapidly since 2014 and peaking in 2024. Collaborative Filtering (CF) remains the dominant approach, but hybrid approaches and deep learning techniques are increasingly being applied to improve accuracy and relevance. The main challenges faced include algorithm optimization, data scarcity, and cold starts, as well as the use of hybrid and deep learning techniques that require more resources. Further research is needed to develop more efficient and personalized algorithms in the digital library recommendation system. Implications: The research offers insights to improve recommendation system efficiency and relevance in digital libraries, addressing key algorithmic challenges. Novelty: This research provides a deeper understanding of recommendation system applications in digital libraries, identifying challenges, future directions, and solutions that combine various algorithms to enhance user experience.]]>
