<![CDATA[ABSTRACT This study examines the comparison between manual library classification using the Dewey Decimal Classification (DDC) system and automated classification assisted by Machine Learning as a response to the increasing volume of library collections and the need for more efficient information organization. The research aims to identify the strengths, limitations, and implementation characteristics of both methods, as well as explore their potential integration in modern library management. Using a Systematic Literature Review (SLR) approach, the study analyzes selected publications that discuss the application of DDC and machine learning algorithms in classifying library materials. The findings show that DDC remains relevant due to its hierarchical, standardized structure and its effectiveness in organizing collections, particularly in small to medium-scale libraries. However, DDC is limited by its dependence on librarian expertise, processing time, and inconsistencies caused by subjective interpretation. In contrast, machine learning offers speed, efficiency, and consistent results when supported by quality datasets, although it has difficulty producing detailed classification numbers and still requires librarian supervision. The study concludes that integrating both methods through a hybrid approach can improve accuracy and efficiency in collection management, supporting libraries in adapting to digital transformation.]]>
Copyrights © 2026
