MORPHOLOGY ANALYSIS FOR MACHINE TRANSLATION

摘要

A method and system for analyzing language morphology to facilitate statistical machine translation from a first language to a target language is disclosed. The method and system can include marking words in a first language text with a part-of-speech tag, and parsing one or more sentences in the first language text to determine syntactic dependence relations between the words in the one or more sentences of the first language text. Morphological features of the words in the first language text can also be extracted. The method and system can also include tagging the words in the first language text with a language morphology tag from a second language. A modified form of words in the first language text can be generated using the language morphology tags from the second language. The modified form of the words can be used to facilitate translation from the first language to a target language.

主权项

1. A method comprising:
marking one or more words in a first language text with a part-of-speech tag; parsing one or more sentences in the first language text, to determine syntactic dependence relations between the one or more words in the one or more sentences of the first language text; extracting morphological features of the one or more words in the first language text; tagging the one or more words in the first language text with a language morphology tag from a second language; and generating, in response to tagging the one or more words in the first language text with a language morphology tag from a second language, a modified form of the one or more words.