=================================================================================

RAKE begins keyword extraction from a document by parsing its text into a set of candidate keywords with the steps below: [1]
          i) The document text is split into an array of words by specified word delimiters.
          ii) This array is split into sequences of contiguous words at phrase delimiters and stop word positions.
          iii) The words, which are assigned the same position in the text and together within a sequence are considered a candidate keyword.

After the candidate keywords are scored, the top T scoring candidates are selected as keywords for the document. [1]

============================================

[1] Michael W. Berry and Jacob Kogan, Text Mining: Applications and Theory, 2010.