Shuangyong Song

Research & Innovation

Linking Images to Semantic Knowledge Base with User-generated Tags

Images account for an important part of Multimedia Linked Open Data, but currently most of the semantic relations between images and other entities are based on manual semantic annotation. With the popularity of image hosting websites, such as Flickr, plentiful tagging information of images makes it possible to automatically generate semantic relations between images and other semantic entities. In this paper, we propose a model for linking images to semantic knowledge base (KB) with user-generated tags of those images, while taking into account topical semantic similarity between tags. The experimental results show that our approach can effectively realize the mentioned aim.

CV

Shuangyong Song received the Ph.D. degree from Institute of Automation, Chinese Academy of Sciences. He is now a Senior Research Fellow with Information Technology Laboratory, Fujitsu Research and Development Center. His current research interests include information retrieval, web/text mining, and natural language processing.