Control and Cybernetics a Method of Hybrid Mt for Related Languages *

The paper introduces a hybrid approach to a very specific field in machine translation — the translation of closely related languages. It mentions previous experiments performed for closely related Scandinavian, Slavic, Turkic and Romanic languages and describes a novel method, a combination of a simple shallow parser of the source language (Czech) combined with a stochastic ranker of (parts of) sentences in the target language (Slovak, Russian, Slovenian). The ranker exploits a simple stochastic model of the target language and its main task is to choose the best translation among those provided by the system. The last section of the paper presents results indicating better translation quality compared to the existing MT system for Czech and Slovak and compares them with the results obtained by the translation from Czech to Russian using the same system.