AutoSPARQL

The aim of the AutoSPARQL is to provide robust question answering over RDF data by combining methods from several research areas, such as:

• natural language processing for creating sophisticated semantic representations of questions
• inductive active learning for incorporating user feedback
• results of the BOA project

The underlying idea is to convert a natural language expression to a SPARQL query, which can then retrieve the answer of a question from a given triple store.

Papers:

• ESWC 2011 AutoSPARQL Paper

Links:

• AutoSPARQL demo
• AutoSPARQL-TBSL demo

Sourcecode:

• main: https://github.com/AKSW/AutoSPARQL
• machine learning algorithms in DL-Learner
• natural language patterns in BOA

Contact

Dr. Jens Lehmann Johannisgasse 26, Zimmer 5-10 04103 Leipzig Tel.:   +49 341 97-32260 E-Mail, Homepage, Research Group

Lorenz Bühmann Johannisgasse 26, Zimmer 5-10 04103 Leipzig Tel.:   +49 341 97 32253 E-Mail, Research Group, Workpage

Dr. Axel-C. Ngonga Ngomo Johannisgasse 26, Zimmer 5-22 04103 Leipzig Tel.:   +49 341 97-32341 E-Mail, Workpage

Daniel Gerber Johannisgasse 26, Zimmer 5-21 04103 Leipzig Tel.:   +49 341 97 32322 E-Mail, Research Group, Workpage