Possible incorrect misclassification of citation style
While looking over the dataset for citation style transitions, we ran into a a journal that had its articles possibly misclassified for the "number" class. This doesn't seem to be the case for the "author" class. For example PMC2771105 correctly classified as a author, but PMC5725689 and PMC5813175 are classified as number where they are really author.
This is for journal: 'Hum Brain Mapp'
.