Joint event
Summer School on Corpus Linguistics
and
Awareness Seminar on Language Technology
Tusnad - Romania
13-26 July 1997
French
version A version of this page is also accessible at URL http://www.racai.ro/Euro97/ann.html
1. Theme of the EUROLAN'97 Summer School
The theme of Eurolan'97 was "Corpus Linguistics" - the use
of language corpora to develop tools and resources for automated language
processing and to test linguistic theories. As well as lectures by distinguished
international faculty members, there were workshops that built upon the
lectures and aimed specifically at stimulating discussion and further collaboration.
Lectures and workshops were organized in three thematic tracks:
Lexicon and Corpora - The
aim of this track was to confront Summer School participants with the state-of-the-art
in using machine readable dictionaries and other on-line linguistic resources
in Natural Language research and with research on extracting these resources
directly from corpora.
The titles of the lectures were:
-
Nicoletta
Calzolari - University of Pisa
-
- Corpus-based lexicography for Language Engineering
-
Jean-Pierre
Chanod - Xerox Grenoble
-
- NLP finite-state technology: two-level morphology, tokenisation,
Part-of-Speech tagging
-
- Incremental Finite-State Parsing
-
Tomaz Erjavec
- Josef Stefan Institute, Ljubljana
-
- An Introduction to SGML
-
Nancy Ide - Vassar
College
-
- Encoding linguistic corpora
-
- Word sense disambiguation
-
Martin Rajman
- Artificial Intelligence Laboratory, Swiss Federal Institute of Technology,
Laussane
-
- Hidden Markov Models for corpus-based NLP
-
Dan Tufis- Romanian
Academy and ICI-Bucharest
-
- Corpus-based morpho-syntactic processing in a multilingual
environment
Discourse Corpus Linguistics -
This track was aimed at presenting a new trend in Natural Language Processing
research that tries to base progress in discourse theories upon intensive
corpus analysis. This workshop followed up on a workshop held at the University
of Pennsylvania in March 1996 on discourse-coding and corpus-based discourse
analysis.
The titles of the lectures were:
Massimo Poesio
- Centre for Cognitive Science - University of Edinburgh
- Evaluating the reliability of dialogue annotation
- Robust processing of definite descriptions
- Coreference and topic tracing on spoken corpora: the
`MapTask project'
Laurent Romary
- CRIN-CNRS, Nancy
- Annotating coreference for the study of discourse phenomena
John Sinclair
- University of Birmingham
- A Corpus-driven Theory of Meaning
Grammar engineering and grammar
learning - This track was focused both on work being done on formalizing
the Romanian language using HPSG in connection with research in formalizing
other Romance languages and on methods for inferring grammars from corpora.
Researchers that announced or begun their work on Romanian HPSG at the
ACM Summer School in Belis-Fintinele, Romania in July 1996, as well as
others researchers, reported their progress within this framework.
The titles of the lectures:
-
Liviu Ciortuz
- University "Alexandru Ioan Cuza" of Iasi
-
- Concurrent parsing with HPSG meta-rules. Application
to Romanian.
-
Aravind
Joshi - University of Pennsylvania
-
- Various aspects of lexicalized grammars, partial parsing,
supertagging and applications
-
Paola
Monachesi - University of Tuebingen
-
- Phenomena in Italian HPSG: clitics, reflexives, clitic
climbing, complex predicates. Relations to other Romance languages (French,
Romanian)
-
Martin Rajman
- Artificial Intelligence Laboratory, Swiss Federal Institute of Technology,
Laussane
-
- Probabilistic methods in corpus-based NLP
-
- Stochastic grammars and their extension to data-oriented
parsing
-
Hans Uszkoreit
- DFKI and University of Saarbruecken
-
- Linguistically interpreted data for language competence
and performance
-
Michael Zock - LIMSI-Orsay,
University of Paris-Sud
-
- Lexical choice and computation of syntactic structure
(pattern matching) from statistics on corpora
2. The Awareness Seminar on Language
Technology
The one day Seminar tried to bring together opinion formers,
decision makers, academic/research and industrial groups, highlighting
the global strategic importance of Language Tehnology for the competitiveness
of business, industries and administration, the tremendous potential that
Language Engineering may have for the information-based society. By exploiting
Information and Language Technologies (spoken and written) the human communication
is enhanced and eased and thereby socio-economic developments are supported
whilst maintaining the diversity of languages and cultures in Europe. Approaching
these groups, and especially opinion formers, and make them realize the
benefits of undertaking a major initiative in the Language Engineering
field, and the risk of not doing it, are some of the key subjects.
3. Venue
Tusnad is a holiday resort situated at about 200 km North
from Bucharest and 67 km from Brasov in the Carpathian mountains. More
about Tusnad you can find here
4. Participants
The Summer School was designed for graduate students and
researchers in Natural Language Processing, including computer scientists
and linguists interested in corpus research.
The Seminar was perceived as an important European event
in the field, being attended by more than 100 participants, goverment and
public decision makers, industry, research/development and university people.
The list of participants and their addresses can be found
here.
5. Background
The Eurolan series of Summer Schools was established in 1993
to stimulate young researchers to progress towards the highest levels of
Natural Language Processing and Language Technology research in their own
countries. The first Eurolan summer
school was held in 1993 in Iasi (Romania),
with the theme "Natural Language and Logic Programming". It was jointly
supported by the French government and the University of Iasi. Seven invited
faculty members gave lectures to 45 students from Russia, Moldavia, Romania,
Bulgaria and Albania. The second Eurolan summer school
was held in 1995,
again in Iasi, with the theme "Language and Perception: Representations
and Processes". This time there were eight faculty members giving lectures
to 55 students from six different countries. The second summer school was
jointly funded by the European Union, the Romanian Ministry for Research
and Development, and the University of Iasi.
6. General information
All lectures were given to all participants. Each of the
three workshops lasted for one day. Each workshop included a student session
in the morning and a round table and/or a working session in the afternoon.
Workshop on Grammar Engineering and Grammar
Learning
Fees for Eurolan'97 were as follows:
-
Tuition: USD 135
-
Accommodation and meals: USD 325
-
Accompanying adult (accommodation and meals): USD 325
-
Accompanying child (accommodation and meals): USD 160.
The fees also included the transport (Bucharest - Baile Tusnad,
on 13 July and Baile Tusnad - Bucharest, on 26 July).
A number of full and partial scholarships were available
for people having difficulties in attending the event. Excepting the TELRI
members, the scholarships didn't cover travel expenses outside Romania.
The full scholarship amounted to USD 460 while a partial scholarship amounted
to USD 410.
The participants were housed in the hotels "Raza Soarelui"
and "Olt" in 2 and 3 beds rooms. Meals were taken in the Restaurant "Veverita".
7. Organization
The Eurolan'97/Awareness Seminar was jointly organized and
co-sponsored by:
-
The European Union - Directorate XIII
-
Association of Computing Machinery (ACM)
-
Association of Computational Linguistics - European Chapter
(EACL)
-
The TransEuropean Linguistic Resources Infrastructure (TELRI)
Copernicus Concerted Action
-
The French Cultural Services of the French Embassy in Bucharest
-
The University "Alexandru Ioan Cuza" Iasi
-
Romanian Academy
-
Romanian ACM Chapter
-
Romanian Ministry of Research and Technology
-
DFKI Saarbruecken
-
Swiss Federal Institute of Technology Laussane.
The Organizing Committee of the joint event express
warm thanks to all our sponsors.
The Program Committee of the joint event:
-
Liviu Ciortuz (University "Alexandru Ioan Cuza", Iasi)
-
Dan Cristea (University "Alexandru Ioan Cuza", Iasi)
-
Paola Monachesi (University of Tuebingen)
-
Massimo Poesio - University of Edinburgh
-
Gheorghe Popa (University "Alexandru Ioan Cuza", Iasi) -
Chair
-
Dan Tufis (Romanian Academy, Bucharest)
-
Hans Uszkoreit (DFKI, Saarbruecken)
-
Grazyna Wojcieszko (European Commission - Directorate-General
XIII)
The Organizing Committee included:
Non-local members
Cristina Peti (POLITEHNICA University, Bucharest)
Dan Tufis (Romanian Academy, Bucharest)
Claus Unger (ACM, University of Hagen)
Hans Uszkoreit (DFKI, Saarbruecken)
Local members (from the University "Alexandru Ioan
Cuza" Iasi)
Dan Cristea
Amalia Todirascu
The accommodation, meals, travel and entertainment were
organized by the NET Agency - Iasi,
Romania.
Last modified by Amalia
Todirascu on 16th September 1998.