'bias': 1.0,

'word.lower()': word.lower(),

#'word.lemma()':lemmatizer(word, postag),

'word[-3:]': word[-3:],

'word[-2:]': word[-2:],

'word.isupper()': word.isupper(),

'word.istitle()': word.istitle(),

'word.isdigit()': word.isdigit(),

'postag': postag,

'postag[:2]': postag[:2],

these features i have considered.

Hi,

is there any benchmark available comparing data size( in terms of number of sentences ) vs performance (in terms of precision , recall, f1 score) ?

why am i asking this ? because when i try with limited dataset ~200 sentences (160-train : 40-test), i am getting max 60% performance.

what alternate i can do to get performance ? or is it limitation of CRF ?

Thanks in advance.