This dataset contains the MASC 3.0 corpus, a large English corpus covering a wide range of genres of written and spoken text, enhanced with semantic annotations, both word...