作者: Timothy Francis O'Donoghue
DOI:
关键词:
摘要: Aligned corpora are generated or received from an external source. Each corpus comprises a set of portions aligned with corresponding the other corpus, for example, so that nominally translations one another in two languages. A statistical database is compiled. An evaluation module calculates correlation scores pairs words chosen each corpus. Given pair text (one language) combines word to obtain alignment score portions. These can be used verify translation and/or modify remove improbable alignments.