Workshop on
CORPUS-BASED NATURAL LANGUAGE PROCESSING
17th Dec 2001-2nd Jan 2002


ANNA UNIVERSITY

CHENNAI


Organised by
The AU-KBC Research Centre, Anna University

Jointly with
The Language Technology Research Centre, IIIT Hyderabad
The National Centre for Software Technology, Mumbai
The School of Computer Science & Engineering, Anna University
&
The Tamil University, Thanjavur

The world is passing through an Information Revolution in which the ability to generate disseminate, access and utilise information is becoming a key determinant of development and progress, for individuals as well as nations. Since  humans express and absorb information and knowledge best through their own ?natural? languages, language processing by machines has come to the centre stage. Example applications are computers translating from one human language to another, humans and computers communicating with each other through written and spoken forms of human languages, etc.

This workshop will introduce the participants to some emerging techniques of machine processing of natural languages based on Language Corpora, with emphasis on Statistical Processing and Machine Translation(MT).

i. Provide an understanding of  NLP in the context of Machine-Translation, Multilingual Information Retrieval and other applications related to Indian Languages and English

ii. Introduce the use of Statistical Techniques on Lexical Resources to refine rule-based methods.These will  be used to develop the NLP applications mentioned earlier.

iv. Identify and define a set of problems to be pursued by the participants in future in the context of  the NLP of Indian languages.

Additionally, faculty from the Organising Institutions would also be contributing to the Workshop through lectures, labs and tutorials.

In addition to the hands-on lab classes, the teams of participants would also be involved in executing  certain real-life Projects as a part of the workshop, broadly towards realising an English-Indian language MT solution, and/or a cross-lingual Information Retrieval  system. Necessary multilingual corpora and other resources are being created ,and is expected to be made available to the workshop ,in Tamil, Hindi and Telugu, in addition to English. The training imparted should enable the participants to pursue state-of-the-art NLP work in their home institutions .

While some familiarity with NLP work  would be most helpful, we would also be admitting a limited number of candidates without any prior exposure to it, with the condition that they undergo a Special Preparatory  Program of three days during the 14th-16th  Dec. conducted at the Wokshop venue. The  background necessary to benefit from the Workshop would be provided through this program.   

There would be a fee of Rs.5000/- per candidate, which would cover the Workshop materials, Tea and Lunch. The candidates have to make their own arrangements for  travel ,stay, breakfast and dinner. Assistance  in arranging  these would however be available from the organisers on prior request, including limited on-campus accommodation. Some fee reduction could be considered under  exceptional circumstances for a small number of candidates on their request, at the discretion of the Organisers.

The application should give full details about the candidate?s  educational  background and experience, with  special emphasis on NLP-related areas, if any. Also to be stated clearly is your computer background,if any. Based on the information provided ,a decision would be taken by the organisers on whether the candidate should go through the Special Preparatory Program during the 14th-16th Dec. or not.