Integrating Commonsense Knowledge into the Semantic Annotation of Narrative Media Objects

In this paper we present an innovative approach for semantic annotation of narrative media objects (video, text, audio, etc.) that integrates vast commonsense ontological knowledge to a novel ontologybased model of narrative, Drammar (focused on the dramatic concepts of ‘character’ and ‘action’), to permit the annotation of their narrative features. We also describe the annotation workflow and propose a general architecture that guides the annotation process and permits annotation-based reasoning and search operations. We finally illustrate the proposed annotation model through real examples.