The dataset consists of PQ scores of Twitter leaders' usernames derived from the metric score, language score, and sentiment score, the whole formula can be found in the report.
