Automatic labelling of data clusters: Alignment checking and unsupervised clustering