Areas

• POS Tagging: assigning an appropriate syntactic class to each of the words in a sentence. Close to human levels of performance can be achieved.
• Machine Translation. Statistical approaches to machine translation trained on bilingual proceedings of the Canadian government.
• Empirical parsing methods based on tree banks (large databases of sentences annotated with syntactic parse trees), such as PCFGS (probabilistic CFGs)
• Word-sense disambiguation, PP attachment, anaphora, discourse segmentation.