Text Wrangling

Which of the following would be of most use in cases where a specific ordering of words is important in a text document?

Correct! N-grams allow for multiple words to be together. Where a bag of words (BOW) uses just single word tokens, phrases can be preserved in an N-gram BOW.

Incorrect. A bag of words includes just single-word tokens and therefore wouldn't allow for information from word ordering to be preserved.

That's not right. A document term matrix (DTM) is essentially a spreadsheet of words (tokens) showing frequency of those tokens in various documents. But word ordering is not preserved in a DTM.

N-grams

Bag of words

Document term matrix