ETAP Linguistic Processor and Construction of a Dependency Treebank for Russian

Igor Boguslavsky
Institute for Information Transmission Problems
Russian Academy of Sciences
Moscow, Russia
e-mail: bogus@iitp.ru

I will present a multi-functional linguistic processor (ETAP) developed in the Institute for Information Transmission Problems of the Russian Academy of Sciences. It can be considered as a computer implementation of the "Meaning - Text" theory by Melchuk and Apresjan. Its major functions include: machine translation, multi-lingual communication via an interlingua (UNL), natural language interface to databases, paraphrasing preserving the meaning, checking and correction of syntactic errors in Russian texts. More details will be given on the types of information entered into the lexicon. They include: lemma name, inflexion type, part of speech, syntactic features, semantic features, subcategorization frame, default translation into all working languages, rules of different types necessary for the processing of the word and lexical functions.

The ETAP system is also used for the construction of a dependency-based treebank for Russian. If needed, syntactic structures provided by ETAP are corrected manually by means of a graphical structure editor.

See previous, next abstract.