Violeta Seretan

Violeta Seretan

Maître-Assistante

Address: 

LATL - Language Technology Laboratory, room L706
Department of Linguistics, University of Geneva
2, rue de Candolle
CH-1211 Geneva, Switzerland
Fly to LATL! [Google Earth link]
Phone: +41 22 379 73 62
Fax: +41 22 379 79 31
E-mail: violeta.seretan@unige.ch

This homepage is no longer maintained. Please visit my new homepage.

I am a Maître-Assistante (Lecturer) at the Department of Linguistics, University of Geneva, and a member of the Language Technology Laboratory headed by Prof. Eric Wehrli. My work is related to two main Computational Linguistics topics, namely syntactic parsing and machine translation, and is focused on the corpus-based acquisition of monolingual and multilingual lexical resources for these applications. I am specifically interested in collocations, a subtype of multi-word expressions, and the interrelation between collocations and parsing/translation. My Ph.D. thesis (defended June, 2008; supervisor Eric Wehrli) explored the use of syntactic information for improving collocation extraction.

Before joining LATL, I studied Computer Science at the University of Iasi, Romania. Both my B.Sc. and M.Sc. theses dealt with Computational Linguistics topics (HPSG, discourse structure, anaphora) and were supervised by Prof. Dan Cristea, who led the NLP Group I was a member of. As a teaching assistant, I currently give practical NLP classes to master students at the Faculty of Arts.

Research Interests

Projects   

I am currently involved in the SNSF project "Analyse multilingue" aimed at the multilingual extension of the Fips symbolic parser (already available for French, English, German, Italian, Spanish, and Greek). I work on the development of the Romanian version, started in a previous related project, "Fips Multilingue" (2004–2006). I previously participated in the "Multra" project (2006–2009) related to the multilingual extension of the Its-2 translation system based on Fips.

I started my work at LATL with the RUIG-GIAN project "Linguistic Analysis and Collocation Extraction" (2002–2004), a joint research with the Translation Division of the World Trade Organisation Geneva, and developed a translation aid tool for collocation extraction and visualisation in parallel corpora. This project was at the root of my PhD (2004–2008).

During my PhD, I spent 3 months as a summer intern at FX-PAL/PARC, Palo Alto, California (August–October 2005). I worked with Lorenzo Thione and Martin van der Berg on interpreting arguments of nominalizations for Question Answering [presentation; talk abstract; Java APIs for NOMLEX].

List of publications
(Peer-reviewed, reverse chronological order, h-index = 8 as computed by "Publish or Perish")


[bib]
Teaching

Reviewing

Conference organisation

Last modified:   novembre 2, 2011