The Papillon Project: The Construction of a Multilingual Lexical Database with a Pivot Structure

Abstract: Papillon is a new R&D project that started as a French-Japanese cooperation between GETA/CLIPS (Grenoble, France) and the National Institute of Informatics (Tokyo, Japan). Its goal is to build a multilingual lexical database and to extract from it digital bilingual dictionaries. The lack of bilingual resources is an obstacle to the development of linguistic software applications, for which adapted dictionaries are a need. As an example, Nippon Telegraph and Telephone in Japan or Lexiquest in France have to develop their own dictionaries in a separate and time-consuming effort. In the academic world, this implies that applications that have been created for French and Japanese offer only a reduced scope, while good English-Japanese pieces of software are available. There is a vacuum to be filled.

In such a context, the GETA/CLIPS and the NII started an R&D project in order to plan and implement a multilingual lexical database. The database development and consultation platform is based on a framework for lexical databases defined by Mathieu Mangeot in his Ph.D. thesis. From the lexical database, it is planned to derive user customized bilingual dictionaries in multiple target formats. It will be possible to generate human usage dictionaries as well as specialized dictionaries for machine translation software. These dictionaries will be available under the terms of an open source license.

This project, initiated by some computational linguists, aims at being useful and open to all those who are interested in English, French, Japanese, Lao, Malay, Thai and Vietnamese. It is also opened to any other language. Moreover, the pivot architecture of the database will facilitate the addition of new languages and save translation efforts.

Enquiries: sals@mri.mq.edu.au