Assess Content Comprehensiveness of Ontologies

This paper proposes a novel method to assess and evaluate content comprehensiveness of ontologies. Comparing to other researchers methods which just count the number of classes and properties, the method concerns about the actual content coverage of ontologies. By applying statistical analysis to a corpus, we assign different weights to different terms chosen from the corpus. These terms are then used for evaluating ontologies. Afterwards, a score is generated for each ontology to mark its content comprehensiveness. Experiments are then appropriately designed to evaluate the qualities of typical ontologies to show the effectiveness of the proposed evaluation method.