iTAG: Easy, Rapid, Automatic Intelligent Tagging for Educational Contents

To enhance the effectiveness of Intelligent Tutoring Systems (ITSs), providing learners with learning objects (LOs) that meet their educational needs is crucial. This requires comprehensive metadata to evaluate the suitability of LOs for personalized learning. Existing metadata models, such as IEEE Learning Object Metadata (LOM), primarily capture bibliographic details but lack the depth needed to describe pedagogical and educational attributes. This paper introduces an intelligent tagger (iTAG), a lightweight, AI-powered platform designed to annotate educational content with rich metadata automatically. By leveraging Natural Language Processing (NLP), machine learning techniques, and domain-specific rules, iTAG automates the extraction of general and pedagogical metadata from text-based LOs. iTAG addresses the limitations of traditional metadata extraction methods by offering higher accuracy, scalability, and resource efficiency, making it suitable for sizeable educational content repositories. The experimental results demonstrate iTAG's ability to accurately generate metadata comparable to expert annotations accurately, achieving 100% precision for basic metadata fields like interactivity type and language and 60-70% accuracy for more complex attributes like semantic density and difficulty. This automatic tagging system offers a practical solution for scaling personalized learning, reducing the dependence on manual metadata annotation while improving the quality of recommendations within ITSs.