Phrase identification in a sequence of words

摘要

Methods and apparatus related to phrase identification. Methods are provided for determining co-occurrence consistencies for positional word pairings of a plurality of sequences of words in a corpus that may be utilized in identifying a phrase; determining a phrase coherence of a sequence of words based on the co-occurrence consistencies for positional word pairings in the sequence of words; and determining one or more phrase boundaries in a sequence of words.