The JSON output contains all the sentences arranged based on their importance. It also contains the total score of individual sentences plus their scores for each features (criteria).

The features used are title, sentence length, and sentence position. More feature / criteria will be implemented soon.