... that training single classifiers and combining their outcomes almost always outperforms a single joint classifier for the three target features In other words, combining the results of G and N (G+N) ... training set We train single classifiers for G (gender), N (number), R (rationality) , GN and GNR, and evaluate them We also combine the tags of the single classifiers into larger tags (G+N, GN+R and ... Learning Features We investigate the contribution of different learning features in predicting functional gender, number and rationality features The learning features are explored in the following...