PhD positions - Computer Science, Natural Language Processing - Marseille, France

The NLP group of the Laboratoire d'Informatique Fondamentale of the Aix Marseille University has 2 open PhD positions in Computer Science in the context of an European project (3 years).

Location : Campus de Luminy (http://sciences.univ-amu.fr/sites-geographiques/site-luminy), Marseille, France
Starting : October 1st 2013
Deadline for application: July 15th 2013

The two fully funded PhD studentships will focus on 2 different aspects of the project:
1- Discourse parsing of speech.
This PhD will focus on developing discourse analysis methods adapted to a large range of conversational styles and domains from spoken conversations to social media interaction. Automatic methods will be studied for five types of discourse analysis: discourse parsing, event and temporal structure, argumentation structure and intra-document coreference. The applicative domain of this research will be the automatic summarization task of human-human dialogs. The language targeted will be French and English.

2- Syntactic and semantic parsing with deep learning methods.
Recently, neural network approaches based on a deep learning paradigm has been successfully applied to some NLP tasks such as POS and NE tagging or dependency parsing ( http://static.googleusercontent.com/external_content/untrusted_dlcp/research.google.com/de//pubs/archive/35671.pdf ). In this PhD we will investigate how this paradigm performs and can be adapted in the context of robust syntactic and semantic parsing of human human conversations collected on social media platforms and telephone call centres.

The successful candidates should :
- hold a relevant degree in the field of Natural Language Processing or Machine Learning.
- have good algorithmic and programming skills

Description of the lab:
The University of Aix-Marseille (AMU) is currently one of the largest university in France, created in 2012 from the fusion of the 3 former Aix-Marseille universities (Université de Provence, Université de la Méditerranée, Université Paul Cézanne).
The LIF (Fundamental Computer Science Lab), is a JRU between the Centre National de la Recherhe Scientifique (CNRS) and AMU. The Natural Language Processing group of LIF aims at developing symbolic and statistical methods for the automatic processing of textual and speech data. The two main characteristics of this research group are: (1) to host both linguists working on the development of rich linguistic resources such as syntactic and semantic lexicons and computer scientists with a very strong experience in numerical approaches for NLP; and (2) to work on both spoken and written languages, at the descriptive level as well as the application level thanks to a strong expertise of some members of the group in Automatic Speech Recognition methods.

Contacts:
Alexis Nasr : alexis.nasr@lif.univ-mrs.fr
Frederic Bechet : frederic.bechet@lif.univ-mrs.fr
Benoit Favre : benoit.favre@lif.univ-mrs.fr