Treebank Statistics: UD_Romanian-SiMoNERo: Features: NumForm
This feature is language-specific.
It occurs with 4 different values: Combi, Digit, Roman, Word.
4572 tokens (3%) have a non-empty value of NumForm.
920 types (5%) occur at least once with a non-empty value of NumForm.
912 lemmas (9%) occur at least once with a non-empty value of NumForm.
The feature is used with 2 part-of-speech tags: NUM (4568; 3% instances), ADJ (4; 0% instances).
NUM
4568 NUM tokens (99% of all NUM tokens) have a non-empty value of NumForm.
The most frequent other feature values with which NUM and NumForm co-occurred: NumType=Card (4126; 90%), Number=Sing (4053; 89%).
NUM tokens may have the following values of NumForm:
Combi(2; 0% of non-emptyNumForm): 31.12.2012Digit(3890; 85% of non-emptyNumForm): 2, 1, 3, 4, 5, 30, 10, 20, 6, 15Roman(182; 4% of non-emptyNumForm): II, iv, III, i, V, l, VII, XIX, I-, VIIIWord(494; 11% of non-emptyNumForm): două, trei, primul, prima, primele, doua, doilea, patru, primă, cinci
NumForm seems to be lexical feature of NUM. 100% lemmas (911) occur only with one value of NumForm.
ADJ
4 ADJ tokens (0% of all ADJ tokens) have a non-empty value of NumForm.
The most frequent other feature values with which ADJ and NumForm co-occurred: Degree=EMPTY (4; 100%), Number=Sing (4; 100%), Case=Nom (3; 75%), Definite=Def (3; 75%), Gender=Masc (3; 75%).
ADJ tokens may have the following values of NumForm:
Word(4; 100% of non-emptyNumForm): ultimul, opta, primul
Relations with Agreement in NumForm
The 10 most frequent relations where parent and child node agree in NumForm:
NUM –[conj]–> NUM (366; 99%),
NUM –[nummod]–> NUM (128; 91%),
NUM –[parataxis]–> NUM (20; 100%),
NUM –[appos]–> NUM (2; 100%).