DFKI-LT - Exploiting the Russian National Corpus in the Development of a Russian Resource Grammar
Exploiting the Russian National Corpus in the Development of a Russian Resource Grammar
2 Workshop on Adaptation of Language Resources and Technology to New Domains,
In this paper we present the on-going grammar engineering project in our group for developing in parallel resource precision grammars for Slavic languages. The project utilizes DELPH-IN software (LKB/[incr tsdb()]) as the grammar development platform, and has strong affinity to the LinGO Grammar Matrix project. It is innovative in that we focus on a closed set of related but extremely diverse languages. The goal is to encode mutually interoperable analyses of a wide variety of linguistic phenomena, taking into account eminent typological commonalities and systematic differences. As one major objective of the project, we aim to develop a core Slavic grammar whose components can be commonly shared among the set of languages, and facilitate new grammar development. As a showcase, we discuss a small HPSG grammar for Russian. The interesting bit of this grammar is that the development is assisted by interfacing with existing corpora and processing tools for the language, which saves significant amount of engineering effort.
Files: BibTeX, W09-41.pdf, ranlp_2009_paper.final-1.pdf