Lexeme Forge (Kuźnia)

Lexeme Forge (Kuźnia) is a tool for collaborative creation of inflectional dictionaries of Polish, developed within the European CESAR project. It is currently used in work on the next edition of the Grammatical Dictionary of Polish (SGJP), from which it inherits its conceptual system. Within CLARIN-PL, the tool was adapted for use by any user with basic knowledge of lexicography as a tool for (co-)creating domain-specific inflectional dictionaries.

Lexeme Forge can export data to the formats of two Polish morphological analysers: Morfeusz and Morfologik. Both analysers are distributed with data from dictionaries of general Polish, which makes it possible to analyse a very large number of documents correctly. However, no dictionary is complete, which is particularly noticeable when analysing specialist or community-specific texts. Lexeme Forge makes it possible to fill this gap by adding missing data from any domain and any register of Polish. Moreover, it allows building one large database in which the vocabulary is divided into separate sets that share a common part and have their own specific parts. Vocabulary from the common part needs to be described only once and can then be attached to as many dictionaries as needed.

Thanks to the possibility of building a complete lexical resource, Lexeme Forge allows, on the one hand, adding data from any domain and any register of the language and, on the other hand, thanks to its internal division into (sub)dictionaries, obtaining precise results by matching an appropriate subset of data to the analysed text.

Registration

Lexeme Forge can be installed on your own server (see below), but you can also use the public installation prepared for CLARIN-PL. The first step to creating your own dictionary from scratch is creating an account. To do so, fill in the registration form at http://kuznia.ipipan.clarin-pl.eu/accounts/register/. An activation link will be sent to the e-mail address provided. After activation, you can log in with your user name and password.

Source code

The most current version of Lexeme Forge may be found on Bitbucket repository.

This page offers the official BSD License release of Lexeme Forge, a tool for collaborative work on inflection dictionaries. By downloading the Lexeme Forge package you accept the conditions of that license.

Principal developer: Jan Szejko Jan.Szejko@ipipan.waw.pl
License: BSD 2-Clause License