Which of the following is not the way to address the curse of dimensionality in text data?
a. Removing punctuations
b. Stemming
c. Adding n-Grams to the dataset
d. Lemmatization