The code below as an example for analyzing massive corpus. I want to restrict the term-document matrix to 1000 most frequent unigrams, but changing the max-features
max-features