Extracting Semantic Transfer Rules from Parallel Corpora with SMT Phrase Aligners
Petter Haugereid, Francis T. Bond · Meeting of the Association for Computational Linguistics · 2012
This paper presents two procedures for extracting transfer rules from parallel corpora for use in a rule-based Japanese-English MT system. First a shallow method where the parallel corpus is lemmatized before it is aligned by a phrase aligner, and then a method where the parallel corpus is parsed by deep parsers before the resulting predicates are aligned by phrase aligners. In both procedures, the phrase tables produced by the phrase aligners are used to extract semantic transfer rules. The procedures were employed on a 10 million word Japanese English parallel corpus and 190,000 semantic transfer rules were extracted.