Hi,
   I wrote a model that classifies tokens by looking at part-of-speech tag
of the surrounding words, for example I want to extract all CD tagged
tokens whose previous token is tagged as NNP and next token is tagged as
NNS. I am not sure of the format of the training data, could someone tell
me how to determine the format of training data when writing our own models
using MaxEnt library.
    Thanks.

Reply via email to