Morphologically and Syntactically Annotated Corpora of Many Languages

Annotated corpora have become a standard resource for research in both linguistics and computational processing of natural languages. Lexicographers judge word usage and distribution by occurrences in corpora; part-of-speech tags may help them narrow their queries. Grammarians may use syntactically annotated corpora (treebanks) for queries such as “show me all examples where a verb governs two objects in the accusative.” In natural language processing (NLP), syntactic parsing is an important preparatory step for many tasks such as question answering, data mining or machine translation; the state-of-the-art parsers rely on human-annotated treebanks and apply machine learning algorithms to extract linguistic knowledge from the treebanks.