2 September: fyi -- phd funding & postdoc, NL
Index of September 2003 | Index of year: 2003 | Full index
Job opening URL: http://ilk.uvt.nl/jobs/ilk2003-imix.html
The Induction of Linguistic Knowledge (ILK) research group of Tilburg
University is seeking a:
* Postdoc, fulltime, two years / vacancy number 810.03.08
* Ph.D. student, fulltime, four years / vacancy number 810.03.09
ROLAQUAD project description
The Robust language understanding for question-answering dialogues
(ROLAQUAD) project goals are to apply advanced machine learning
techniques to a class of "shallow understanding" tasks in a
question-answering dialogue system. Such a system will have several
information threads running and changing in parallel through time:
which moves did the user made so far and what type is the current move?
What has the user asked for, generally and specifically? What answers
have been given and what answer should best be given next? These
threads are noisy - they are based on noisy input (e.g. from speech
recognition), and contain partially and mutually uncertain
classifications and decisions. Advanced machine learning methods can be
used to optimize over joint uncertainties and resolve ambiguities at
multiple levels; a type of task for which standard classification-based
algorithms are not typically suited.
In the project, state-of-the-art machine learning algorithms will be
implemented (a.o., maximum- entropy markov models and conditional
random fields). The algorithms will be applied to annotated data of a
domain that is yet to be determined by the IMIX project as a whole
(consisting of groups from the universities of Groningen, Nijmegen and
Tilburg). Throughout the project, there will be close collaboration
with all groups and integration of modules developed at these groups
(speech recognition, parsing, multi-modal generation).
Funding for this research has been made available by the Dutch
Organisation for Scientific Research (NWO), as part of their new IMIX
(Interactive Multi-modal Information eXtraction) Programme. Project
supervisers are prof. Walter Daelemans and dr. Antal van den Bosch. The
project will be embedded in the ILK Research Group and will be
supported by a scientific programmer.
[1] Postdoc position (two years)
The postdoc will be responsible for the development of the prototype
QA-understanding module, which integrates semantic question parsing,
dialogue move detection, dialogue topic detection, and move generation.
Dynamically, this ranges from the development of submodules based on
standard classification-based machine learning algorithms and existing
NLP applications (shallow parsing), via the integration of these
modules into a first prototype, to the expansion of this prototype
towards the integration of new modules based on advanced machine
learning algorithms.
Applicants should have a Ph.D. degree in computational linguistics,
cognitive science, or computer science, and a background in corpus
linguistics, statistical NLP - in particular question answering, and/or
machine learning. Programming experience (C/C++/Perl/Java) is a pro.
Applicants should have, or be willing to develop, active knowledge of
spoken and written Dutch, which is the target language within the IMIX
project.
The collective labour agreement of Tilburg University applies. The
selected candidate will get a full time two year contract. The postdoc
position has a gross monthly salary that depends on experience, with a
minimum gross monthly salary of EUR 2.136,-- and a maximum of EUR
3.694,--. The target starting date is flexible from November 1, 2003.
[2] Ph.D. position (four years)
The Ph.D. student will work on a four-year project plan that culminates
in the writing and defense of a Ph.D. thesis. In the first two years,
the Ph.D. student cooperates with the postdoc and the scientific
programmer on the development of a prototype QA understanding module.
Starting from a baseline system composed of existing
classification-based machine learning modules, the Ph.D. student
designs the architecure of the second-stage QA understanding module in
which advanced machine learning algorithms manage the multi-threaded
handling and analysis of information
streams. Subsequently, the Ph.D. student performs series of experiments
on the developed system to establish sound empirical estimations of
generalization performance aspects of the system.
Applicants should have an honours / Master's degree (or equivalent) in
computer science, computational linguistics, or cognitive science.
Programming experience (C/C++/Perl/Java, applications, system design)
is required. A background in corpus linguistics, statistical NLP (in
particular question answering), and/or machine learning is a pro.
Applicants should have, or be willing to develop, active knowledge of
spoken and written Dutch, which is the target language within the IMIX
project.
The selected candidate will get a four year contract (full-time) as a
Ph.D. student. The full-time Ph.D. position has a gross monthly salary
of EUR 1.611,-- in the first year, increasing to EUR 2.185,-- in the
fourth year. The target starting date is November 1, 2003.
For more information on both positions, contact Antal van den Bosch via
email (Antal.vdnBosch@uvt.nl) or phone (+31.13.4663117). Applications
for both positions should
include cover letter, CV, and names of two references, and should be
sent before the application deadline of September 15, 2003 by regular
mail to
Drs. O. Zweekhorst
Managing Director, Faculty of Arts, Room U 40
Tilburg University
P.O. Box 90153
NL-5000 LE Tilburg
The Netherlands
or by email to solliciterenFDL@uvt.nl . In your letter, either make
reference to vacancy code 810.03.08 for the postdoc position, or to
810.03.09 for the Ph.D. position.
It is the policy of the university to increase the number of women
employed. Also members of ethnic minority groups and handicapped people
are expressly invited to apply.
Tilburg University is a modern, specialized university in the field of
the social sciences,
with more than 9,000 students and more than 1,600 staff. The education
and research at the university meet nationally recognized high
standards. In its education and research, the Faculty of Arts
addresses the following fields of interest: text linguistics and
communication, professional communication and digital media,
linguistics and artificial intelligence, literary studies and consumer
behaviour with respect to literature. The faculty participates in the
inter-university Research Institute Centre for Language Studies, and is
associated with the National Research School of Linguistics, the
Research School of Logic, and the Research School of Literary Studies.
Characteristics of the faculty are its small scale, its social
orientation, and the high quality of its education and research.
Index of September 2003 | Index of year: 2003 | Full index