Exploiting Parallel Corpora for Supervised Word-Sense Disambiguation in English-Hungarian Machine Translation

In this paper we present an experiment to automatically generate annotated training corpora for a supervised word sense disambiguation module operating in an English-Hungarian and a Hungarian-English machine translation system. Training examples for the WSD module are produced by annotating ambiguous lexical items in the source language (words having several possible translations) with their proper target language translations. Since manually annotating training examples is very costly, we are experimenting with a method to extract examples automatically from parallel corpora. Our algorithm relies on monolingual and bilingual lexicons and dictionaries in addition to statistical methods in order to annotate examples extracted from a large EnglishHungarian parallel corpus accurately aligned at sentence level. In the paper, we present an experiment with the English noun state, where we categorized its different occurrences in the Hunglish parallel corpus. Our experiment showed that 93% of all corpus occurrences of state formed multiword lexemes with unambiguous Hungarian translations, hence these can be omitted from the training data. The remaining 7% of all occurrences is still sufficient for producing training data.