Application of Artificial Intelligence and Machine Learning in Libraries: A Systematic Review

Abstract

Academics, researchers, and information professionals are all actively involved in AI and ML research because of the growing
importance of these concepts and their applications. The purpose of this research was to compile findings from empirical studies
that investigated the use of AI and ML in library settings. A systematic literature evaluation was carried out in accordance
with the initial criteria supplied by Kitchenham et al. (2009) in order to accomplish the study’s aims. The databases Scopus,
LISA, and LISTA as well as the Web of Science were used to gather data. A total of thirty-two publications were ultimately chosen,
examined, and analyzed after following the strict/established selection method. The most common domains and methods
utilized in AI and ML were the subject of thirty-two publications that were examined and summarized. The results demonstrate
that theoretical works constitute the majority of AI and ML research in the LIS sector at the present time. On the other hand,
case studies and implementation projects were also the focus of other scholars. Several machine learning approaches, such as
logistic regression, KNN, and AdaBoost, have found extensive usage in library collection management for tasks such as metadata
production, resource identification, and book procurement. While recommender systems, support vector machines, and
association rules have been used for circulation-related tasks such as book suggestion, user rating, and bibliographic data. In
order to promote more tech-oriented methods and to anticipate future innovation routes, this study will provide academics,
practitioners, and educators a bird’s-eye perspective of AI and ML in libraries.