Skip to content

Latest commit

 

History

History
28 lines (18 loc) · 597 Bytes

README.md

File metadata and controls

28 lines (18 loc) · 597 Bytes

nlp_spam_detection

Natural Languege Processing

TODO

nltk tf-idf countvector regex feature engineering

picture of body length

picture of body punct%

cross validation grid search machine learning algorithms RF and GB

Enference

model/score Fit time Predict time Precision Recall Accuracy
Random Forest 4.375 1.384 0.984 0.814 0.972
Gradient Boost 781.057 0.376 0.901 0.872 0.969