Prof. Dr. Jens Lehmann

Navigation: Info Pages → Question Answering

Question Answering

Question answering is the area of computer science concerned with automatically answering questions posed by humans in natural language. In our research, we investigate closed domain question answering against a local knowledge graph. More specifically, we convert natural language queries to SPARQL queries, which can then be processed and the answers returned to the user.

For achieving this goal, we research two approaches. First, we investigate the analysis of questions via principles of computer linguistics. In that case, there is usually a pipeline of processing steps sequentially transforming the input question until a SPARQL query is obtained. The second line of research we are pursuing are end-to-end question answering systems directly transforming an input question into a SPARQL query. Those systems rely on machine learning methods, e.g. neural networks, and the availability of (usually large amounts of) training data.

Journal Publications

Conference Publications

Poster Articles

Demos

print icon© 2001 - 2017, Jens Lehmann