generalized-language-modeling-toolkit icon indicating copy to clipboard operation
generalized-language-modeling-toolkit copied to clipboard
verified publisher icon renepickhardt

FalseMaximumLikelihood should work for Marginal

Open lschmelzeisen opened this issue 10 years ago • 1 comments

You told me FalsemaximumLikelihood-estimation (namely doing c(a b c)/c(a b) instead of c(a b c)/c(a b _)) should work in the marginal case. It doesn't though. We should research why.

lschmelzeisen avatar Jan 06 '15 18:01 lschmelzeisen

we should do a backoff if c(a b) was not seen (if we do proper backoffs this should add to one but it is not MLE)

renepickhardt avatar Jan 13 '15 14:01 renepickhardt