Peer review process is a cornerstone of scholarly publishing, playing a pivotal role in ensuring the quality and credibility of academic articles. With the increasing volume of scientific publications and the growth of interdisciplinary research, identifying qualified reviewers has become a significant challenge for editors. To address this issue, scholarly reviewer recommendation systems have been developed, leveraging textual data, research records, collaboration networks, and intelligent algorithms to automate and optimize reviewer selection. This study aims to provide a scoping review of the methods and algorithms used in designing these systems and to propose a conceptual framework for their future development.
A comprehensive scoping review was conducted using international databases (IEEE, PubMed, Scopus, Web of Science) and Persian databases (Magiran, SID, IranDoc, Noormags) for the period 2010–2025. After screening, 28 eligible studies were selected and analyzed according to technical approaches, system types, application domains, data sources, reviewer assignment criteria, and evaluation metrics.
The review revealed that combining multiple technical approaches—such as natural language processing for semantic matching between articles and reviewer profiles, network analysis to detect conflicts of interest and hidden relationships, machine learning and author-topic matrix models for expertise classification, and fairness-oriented optimization algorithms for balanced workload distribution—yields the most effective performance in reviewer recommendation systems. Multi-source hybrid systems that integrate textual data, bibliometric indicators, and collaboration networks provide enhanced performance and broader interdisciplinary coverage.
Integrating diverse technical approaches, multi-source data, and semantic and network-based analysis can significantly improve the efficiency and reliability of reviewer recommendation systems. Nevertheless, challenges such as the lack of standardized datasets, difficulties in integrating multiple information sources, and concerns regarding algorithmic transparency and fairness persist. The proposed conceptual framework aims to guide the development of next-generation systems capable of supporting interdisciplinary reviews, ensuring fairness, maintaining ethical standards, and ultimately enhancing the quality, trustworthiness, and advancement of scholarly research.