The AKSW papers “Sorry, I don’t speak SPARQL – Translating SPARQL Queries into Natural Language” and “Question Answering on Interlinked Data” were accepted as full research papers for presentation at WWW2013. The first paper presents SPARQL2NL, a framework which deals with the translation of SPARQL queries (and RDF triples) into natural language. The intention behind SPARQL2NL is to support for lay users who want to know what happens in the backend of the semantic system they are using. The framework has already been integrated into two pieces of software, i.e., the question answering engine TBSL and BioASQ‘s annotation tool for bio-medical questions. Moreover, it was shown to be of help for both SPARQL experts and non-experts. Check out the demo here. The second paper focuses on keyword search and question answering on interlinked data sets and shows how links across knowledge bases can be used to answer questions that require aggregating knowledge from several knowledge bases. The final versions of the papers will be made available soon. Please check them out and let us know what you think.
Looking forward to seeing you at WWW,
Axel for AKSW