This dataset contains morpheme dictionary of the Belarusian language.
The dictionary uses a surface segmentation approach, i.e. the original form of the morphemes is not restored ("funniest" = "funn-i-est").
Each morpheme is labeled with morpheme type (ROOT, PREF for prefix, SUFF for suffix, END for ending or LINK for linking vowels).
In total, the dictionary contains segmentations for 35,219 words.
The dictionary mentions 6,350… See the full description on the dataset page:
https://huggingface.co/datasets/ruscorpora/morphodict-bel.
