Automatic identification of cited text spans: a multi-classifier approach over imbalanced dataset · Study