im working with a corpus that contains 180 documents of movie reviews written by two reviewers. Each document is a review on one movie written by one reviewer. The first 80
