Application of Artificial Intelligence and Machine Learning in Digital Archive Collections in Public University Libraries in South-South Nigeria

Abstract

This study investigated the application of Artificial Intelligence (AI) and Machine Learning (ML) in digital archive collections in public university libraries in South-South Nigeria, focusing on how feature engineering, model training, and performance evaluation influence archival efficiency. Guided by the Technology Acceptance Model (TAM), the study adopted a correlational research design and involved 50 librarians from five public universities, using a structured four-point Likert questionnaire administered via Google Forms. A total of 43 responses were analyzed using regression techniques at the 0.05 significance level. Findings showed that feature engineering, model training, and performance evaluation significantly enhance the effectiveness of digital archive systems. The study concludes that AI/ML techniques improve metadata accuracy, automate classification, and promote user satisfaction in digital repositories, and recommends the adoption of secure data governance, standardized metadata protocols, scalable infrastructure, and sustained funding to optimize AI/ML integration in public university libraries.