"No one is harder on a talented person than the person themselves" - Linda Wilkinson ; "Trust your guts and don't follow the herd" ; "Validate direction not destination" ;

October 29, 2017

Day #77 - Quick Summary - Kaggle Lessons - Features, Dates, Text

  • For Features - One Hot Encoding, Label Encoding, Frequency Encoding, Ranking, MinMaxScaler, StandardScaler
  • For Dates - Periodicity - Year, Date, Week, Time Slice - Time past since particular moment (before / after), Difference in Dates (Datetime_feature1 - Datetime_feature2), Boolean binary indicating date is holiday or not
  • For Text - Preprocessing - Lowercase, Stemming, Lemmatization, stopwords removal, Ngrams can help use local context, Postprocessing - TFiDF,  Use BOW for Ngrams
Happy Coding and Learning!!!

No comments: