use a different word tokenizer that doesn't require external data to be downloaded #1
Description
The NLTK's word_tokenize function requires Punkt data to be downloaded, which could effectively break the program for those that don't know what's going on.