POS Tagging
Penn Treebank POS Scheme
This is a classification scheme of linguistic types with 45 tags for tokens and punctuation schemes
Applications
Extracting Noun Phrases
Math-Aware POS Tagging
Some tweaking is needed to get POS tagging to work with formulas
- main idea: replace Math formula tokens with some hash prepended with “MATH_” or “FORMULA_” and then apply POS Tagging