An Intelligent Framework for Information Organization in Digital Libraries Using Machine Learning on Azure Cloud

This study proposes the Intelligent Library Management System (ILMS), a cloud-native framework designed to enhance information organization and retrieval in digital libraries through the integration of machine learning techniques. The ILMS addresses the limitations of traditional Library Management Systems (LMS), which rely primarily on bibliographic metadata, by enabling automated extraction, structuring, and semantic indexing of unstructured content from PDF and image-based resources. The system was developed using ASP.NET Core API and Blazor WebAssembly, deployed on Microsoft Azure Cloud, and evaluated on a dataset of 638 digital resources comprising 200 eBooks and 438 academic papers spanning 42 subject categories. Machine learning components, including a fine-tuned DistilBERT classifier and a hybrid collaborative filtering recommendation engine, were employed for content classification, extraction, and semantic analysis. Experimental evaluation involved 126 participants over four weeks of usability testing, with load testing conducted across 50 to 1,000 concurrent users using Apache JMeter 5.6. The system achieved a classification accuracy of 89.5%, a text extraction accuracy of 96.3%, and a metadata extraction accuracy of 91.8%. End-to-end search latency was reduced by 74.5%, from 47 seconds to 12 seconds, while recommendation acceptance rates improved from 22% to 34%. The system attained a System Usability Scale (SUS) score of 84/100 and maintained 99.8% availability with stable performance up to 500 concurrent users. These findings demonstrate that integrating machine learning with cloud infrastructure provides a scalable and effective solution for managing large and heterogeneous digital library collections.