I am a research scientist at the SFB 632 "Information Structure" affiliated with the Applied Computational Linguistics Lab in Potsdam (Golm), Germany. My research is driven by the desire to better understand natual language and detect its underlying patterns. The focus of my research in the past years was mainly on realization of salience in discourse, ontologies of linguistic annotations, and multi-layer annotations.

I hold a MSc in Computer Science (Dipl.-Inform.) and MA in Linguistics, both from TU Berlin. In July 2010 I defended my PhD thesis (Dr. Phil.) "Mental Salience and Grammatical Form" supervised by Prof. Manfred Stede and Prof. Anke Lüdeling.

My vision for the next few years involves the integration of linguistic analyses, multiple levels of content and knowledge sources by means of Semantic Web technologies. By doing so, the potential of ontology-based algorithms can be unleashed, so we can get a grasp on more abstract levels of the meaning of language, including discourse semantics, discourse structure, pragmatics and textual inference.

Part of this vision may become reality in the course of on-going community efforts, as manifested, for example, in large-scale research projects such as LOD2. In order to coordinate and to facilitate such activities, we founded the Open Linguistics Working Group, hosted by the Open Knowledge Foundation (OKFN), an interdisciplinary group of enthusiasts working towards building bridges between the Semantic Web community and linguistics. Extending my work on data models for multi-layer annotations (PAULA) and ontologies of linguistic annotations (OLiA), richly annotated corpora can not only be represented in a consistent and standardized way, but using OWL and RDF they can also be linked with lexical-semantic resources from the LOD cloud, as illustrated here with the current draft of the Linguistics LOD cloud diagram. (Click for more information.)

In this context, the PAULA data model and the OLiA ontologies may come to fulfil new and unforeseen roles: PAULA (or its OWL/RDF linearization POWLA) provides a scheme for the representation of arbitrary corpora in OWL/RDF, OLiA mediates between the annotations of these corpora on the one hand and terminology repositories such as the General Ontology of Linguistic Description (GOLD) and the ISO TC37/SC4 Data Category Repository (ISOCat) on the other hand.