Semantic Gap Detection in Metadata of Adaptive Learning Environments

Quality of learning objects metadata, in many respects, defines the quality of an adaptive learning environment presenting these learning objects to a student. Metadata inconsistencies and gaps may be the cause of various problems: from a system malfunction to ineffective learning experiences. In this paper, we propose an intelligent and rigorous mechanism for detecting metadata gaps in collections of learning content. The mechanism converts learning objects metadata into an OWL2 ontology, detects logical conflicts using Semantic Web reasoning techniques and generates human-readable explanations for an author to resolve the gaps. The evaluation of the developed semantic gap detection tool with real learning content collections demonstrates its effectiveness.