On Linked Open Data ( LOD )-based Semantic Video Annotation Systems