search

ai-se / SMOTE

0 stars 0 forks source link

compare to our "baseline" #16

Closed WeiFoo closed 8 years ago

WeiFoo commented 8 years ago

l2 normalization hashing trick linear sum

and what ever else zhe says

WeiFoo commented 8 years ago

l2 normalization is trick to process data to normalize matrix and have to be used in both hashing and tidf.

hashing trick vs. tfidf

linear sum vs [NB_naive, NB_SMOTE, NB_TunedLearner, NB_TuneSMOTE].

results are pending.... up to Jan 27