Introduction

From various projects within our institute it has become clear that with respect to lexica a wide variety of formats, structures and terminologies are being used. These derive for example from the specific purpose for a lexicon and the tools which are being used to create these lexica. However, this makes cross lexica search, merging, linking and comparison an extremely difficult task. These problems have been recognized within the ISO TC37/SC4 group. Work is in progress to provide a common model for the creation and use of very large scale lexical resources and to manage the exchange of data between and among these resources. The result is a flexible model, Lexical Markup Framework (LMF) which allows for customized lexicon creation in a common framework. LEXUS is based upon the LMF model and is the first test and reference implementation of LMF.

LMF Core model

The flexibility of the LMF model means that users may extend the model at any level to suit their specific lexicon task. To incorporate morphology for example a user may define a morphology extension and choose to connect it to any of the components of the core model. Of course, it wouldn't make much sense to connect it to the LexiconInformation component, but the model places no restrictions on the type of components used and the location in the model where they are specified. In general, it encompasses a Lego building block idea where components and data categories may be added to create ones own structure. A number of standard extensions are being defined within the LMF community to assist users by providing a number lexical sub structures reflecting best practices which may be used as a starting point for lexicon creation.

LEXUS also provides support for Data Category Registries. Tag names between lexica may be different although referencing the same linguistic concept. Tag names may be the same, but referring to the different concepts or different vocabularies may be used although the definition is the same. LEXUS attempts to tackle these problems by providing a number of Data Catagory Repositories (including ISO 12620 and Shoebox's MDF) stimulating the user to make use of predefined and accepted concepts and tag names. Of course it is also possible for users to define their own data categories outside of the scope of the provided DCR's.

LEXUS requirements

More indepth information about this project is available on the documentation page. The development team of LEXUS encourages people to take part in defining the requirements for LEXUS. You can get informed about the current status of the project and a shortlist of user requirements is listed here. This list contains the main issues tool development is currently focussing on. Please feel free to inform us of your comments and suggestions.