Extracting Concepts

less than 1 minute read

This article collects some thoughts on normalizing phrases to concepts.


  • drive_car <- "drive a car", "you drive your car", "driving cars" and "drive there in a car",
  • buy_christmas_present <- "I bought a lot of very nice Christmas presents", "let buy christmas presents", ...

  1. POS Tagging -> remove all POS tags not in ('nouns', 'verbs')
  2. lemmatization
  3. combine words with a "_" :-)