Multi Labeled Depression Corpus of 57,000 English tweets

This is a new English corpus for depression detection from Twitter which contains negation words. This corpus consists of 57,391 tweets classified into 24,662 of 1 which means \"depressed\" tweets, 11,923 of 0 which means \"indifferent\" tweets, and 20,806 of -1 which refers to \"happy\". the fetched tweets covered a wide range of English synonymous with depression and happiness on Twitter including negation words . in addition, It has been labelled using the Vader lexicon. finally, this corpus's text has been cleaned and lemmatized.