AutoSPARQL: Convert a natural language expression to a SPARQL query
The underlying idea of the project 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.
The aim of 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
Sourcecode
- main (is not actively maintained anymore)
- machine learning algorithms in DL-Learner
- natural language patterns in BOA
Project Team
- Lorenz Bühmann
- Dr. Konrad Höffner
- Prof. Dr. Jens Lehmann (Principle Contact / Maintainer)
- Prof. Dr. Axel-C. Ngonga Ngomo