Welcome to my homepage.

My name is Behrang QasemiZadeh and am a postdoctoral researcher at the Computational Linguistics department of the University of Duesseldorf.
In this role and under the supervision of Prof. Laura Kallmeyer, in the context of DFG Collaborative Research Centre 991, I investigate the use of data-driven methods for 'hierarchical frame induction'. See the project website here.

Previously, I have worked on several interdisciplinary projects involving analysis of various natural language structures: distributional semantics, information extraction, natural language dialogue systems, text-to-speech, and profiling authors' attributesa are amongst them.


From a computational perspective, my current research deals with 'geometric' machine learning techniques applied to vector spaces that are projected randomly onto low-dimensional spaces, to address the problem of sparse and heavey-tailed distribution of evidences when procesing natural language.


I did my graduate study in Computer Science at the Insight Centre for Data Analytics at the National University of Ireland (Ph.D. 2015), advised by Siegfried Handschuh (and before him Paul Buitelaar). I had the opportunity to work with Ian O'Neill and Philip Hanna from Queen's University, Belfast. I also obtained a Master's degree in Artificial Intelligence from Iran University of Science and Technology, where I worked under the supervision of Mohammadreza Kangavari.

I collaborate with Esther Seyffarth (on the frame induction project), the PARSEME's COST action members (shared task v 1.1 on VMWEs), and Aurelie Herbelot (PeARS distributed search engine). Also, I am involved in organizing the upcoming Task 7 of SemEval 2018 (Semantic Relation Extraction and Classification in Scientific Papers) together with Anne Schumann and Kata Gabor.

I worked for a few companies in the past years, among them are Relative Insight, Digital Clone, and Artificial Intelligence Rayvarz.

• • You can read my cv here (PDF format).
    
    
    
    

NEWS

• SemEval 2019 task 2 on Unsuperivsed Lexical Frame Induction
• Check PARSEME shared task on automatic identification of verbal MWEs - edition 1.1
• Results from SemEval 2018 task 7 on Semantic Relation Extraction and Classification in Scientific Papers are now available.
• A demo of the universal part-of-speech tagger for Parsi (aka Farsi and Persian) is now available online.
• ACL RD-TEC 2.0, a corpus of annotated terms in context, is now available for download.

Dictionaries do not exist to define, but to help people grasp< (Bolinger, 1965) share meanings.
Privacy Statement