Semantic annotation with Pundit: A case study and a practical demonstration

This paper discusses the application of Pundit, a novel semantic annotation tool, in the case study of Burckhardt's correspondence. In this context the occurrences of persons and places names as well as references to works of art in the letters are being semantically annotated and linked to the Web of Data. Such an experiment has a two fold goal. The first one is validating and iteratively improving Pundit as a tool for creating and making explicit the information "hidden" in the letters. The second is to investigate how such knowledge can be presented and visualized to be actually useful for scholars themselves. Leveraging on Semantic Web technologies and on the Open Annotation data model, scholars' collaboratively created annotations are coherently merged with the metadata already present in the DL and exposed via REST APIs to build a number of alternative ways of visualizing the knowledge graph. We claim this approach fosters a "virtuous circle" where the new knowledge produced by scholars can, in turn, become the starting point for new researches. This paper discusses the preliminary results of this ongoing project, presenting requirements and a first visualization prototype. © 2013 ACM.