With the rapid growth of electronic resources, institutional repositories, remote-access platforms, learning management systems, and hybrid academic services, academic libraries are no longer limited to catalogues for discovering information but have been transformed into intelligent knowledge environments. In this shift, knowledge organization systems (KOS) powered by artificial intelligence (AI) are becoming more common to assist with semantic indexing, automated metadata enrichment, natural language search, recommendation and user-centred discovery. This research paper will transform a comparative draft investigation into a journal paper format and compare the use of AI-based semantic retrieval with traditional keyword-based retrieval in a hybrid digital academic library setting to determine if this method provides any benefit in terms of retrieval efficiency and user experience compared to current keyword-based retrieval techniques. It is suggested that a comparative mixed-method design be used, which incorporates elements of the controlled retrieval-task testing, user responses to questionnaires, analysis of the system logs, and librarian input. The efficiency of retrieval will be measured using the following measures: Precision@10, Recall@10, Mean Reciprocal Rank, normalized Discounted Cumulative Gain, average search time, query reformulation, click-through rate, and task success. The results show that the AI-driven semantic retrieval environment has superior performance in terms of accuracy, ranking quality, search effort, user satisfaction, and user trust compared to the keyword-based system. Precision@10 rose from 0.62 to 0.84, Recall@10 rose from 0.55 to 0.79, query reformulation dropped by 50% and average search time dropped from 6.8 minutes to 4.1 minutes. Users' behavior also changed in the direction of longer natural language queries, fewer repeated queries, lower rates of abandonment, and higher intent to re-use. The research shows that the use of AI-powered semantic knowledge organization has the potential to revolutionize academic libraries into dynamic and user-centric discovery spaces, but this transformation can only happen when concerns about metadata quality, ontology-based indexing, digital literacy, privacy concerns, explainability, bias monitoring, and institutional preparedness are addressed.
Naksh (Sat,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: