Good accuracy but low precision and recall in Imbalanced text dataset [closed] The text dataset is extremely imbalanced where out of total 1200 records, minority dataset is around 120. Dataset has two classes 'Y' and 'N'. I have tried Data preprocessed. TfIdfVectorizer with n_grams to achieve more meaningful data. RandomForestClassifier, XGBClassifier, SVC. UpSampling and DownSampling Still I don't get good precision or recall in any of the above methods. I was expecting UpSampling should work fine in this case but it didn't. Am I missing anything? Any suggestion...