Argumentation: mining and linguistic foundations
While automatically identifying and extracting argumentative portions in natural language text has gained momentum, I am particularly interested in the linguistic foundations of arguments in real, multi-party dialogue -- which linguistic devices are used to convey them, how can we categorize them and what automatic methods are best to extract them. This line of research is one of the foci of the ADD-up project (http://arg.tech/add-up), funded by the VolkswagenStiftung from 2017 to 2021.
Visual Analytics in Linguistics
An exciting field in (computational) linguistics is the exploration and interpretation of linguistic information with Visual Analytics systems (LingVis). In cooperation with the visualization group of Prof. Dr. Daniel Keim, I have worked on visualizing semantic change, argumentation and clustering results from machine learning approaches. Projects include the following:
I'm also a co-organizer of the LREC VisLR workshops (Visualization as Added Value in the Development, Use and Evaluation of Lexical Resources), with the first edition at LREC 2014 and the second edition at LREC 2016.
Syntax and semantics
Another interest lies in the theoretical and computational linguistic analysis of Urdu, an Indo-Aryan language mostly spoken in Pakistan. I'm particularly interested in theoretical investigations of the syntax-semantics interface and its representation for NLP systems.