My research interests include:
- Statistical dependency parsing of German and Dutch.
- Parsing and generation using unification grammar.
- Topological field modeling.
- Machine learning (feature selection, linear classifiers, MLPs, RNNs).
Since September 2014 I work as a researcher and lecturer at the Department of General and Computational Linguistics of the University of Tübingen. Starting July 2017, I am co-Principal Investigator of the SFB 833 A3 project. I currently work on statistical dependency parsing.
From August 2013 to September 2014, I worked as a researcher at the Department of General and Computational Linguistics of the University of Tübingen in the CLARIN-D project. Here I worked on improving the scalability of WebLicht and TüNDRA and improving various natural language processing compontents.
From January 2013 to July 2013, I worked as a developer of language technology at GridLine BV. Here, I implemented a new lemmatizer for unknown words, a named entity recognizer, and a lexical resource database. I also designed and implemented a revised architecture for the GridLine spell checker.
From September 2008 to December 2012, I was a PhD candidate at the University of Groningen in the STEVIN DAISY project of the Dutch Scientific Foundation. Here, I worked on natural language generation, machine learning, and error mining under the supervision of prof. Gertjan van Noord.