Treebank Statistics: UD_Polish-PDB: Features: NumForm
This feature is language-specific.
It occurs with 3 different values: Digit, Roman, Word.
4662 tokens (1%) have a non-empty value of NumForm.
832 types (1%) occur at least once with a non-empty value of NumForm.
748 lemmas (3%) occur at least once with a non-empty value of NumForm.
The feature is used with 3 part-of-speech tags: NUM (2633; 1% instances), ADJ (1159; 0% instances), X (870; 0% instances).
NUM
2633 NUM tokens (100% of all NUM tokens) have a non-empty value of NumForm.
The most frequent other feature values with which NUM and NumForm co-occurred: Number=Plur (2575; 98%), Gender=Masc (1853; 70%), Animacy=Inan (1332; 51%).
NUM tokens may have the following values of NumForm:
Digit(1277; 48% of non-emptyNumForm): 10, 3, 2, 30, 5, 15, 20, 4, 50, 12Roman(1; 0% of non-emptyNumForm): IXWord(1355; 51% of non-emptyNumForm): dwie, dwa, dwóch, trzy, trzech, cztery, pięć, dwaj, czterech, pół
NumForm seems to be lexical feature of NUM. 100% lemmas (407) occur only with one value of NumForm.
ADJ
1159 ADJ tokens (3% of all ADJ tokens) have a non-empty value of NumForm.
The most frequent other feature values with which ADJ and NumForm co-occurred: Aspect=EMPTY (1159; 100%), Polarity=EMPTY (1159; 100%), VerbForm=EMPTY (1159; 100%), Voice=EMPTY (1159; 100%), Degree=Pos (1158; 100%), Number=Sing (1130; 97%), Gender=Masc (971; 84%), Animacy=Inan (938; 81%), Case=Gen (668; 58%).
ADJ tokens may have the following values of NumForm:
Digit(1017; 88% of non-emptyNumForm): 1, 2008, 2000, 2009, 2, 15, 1995, 20, 1996, 1997Roman(142; 12% of non-emptyNumForm): II, XIX, I, IV, III, XVI, XVIII, XX, VIII, XII
NumForm seems to be lexical feature of ADJ. 100% lemmas (271) occur only with one value of NumForm.
X
870 X tokens (82% of all X tokens) have a non-empty value of NumForm.
The most frequent other feature values with which X and NumForm co-occurred: Foreign=EMPTY (870; 100%).
X tokens may have the following values of NumForm:
Digit(858; 99% of non-emptyNumForm): 2, 1, 3, 4, 5, 6, 7, 8, 14, 9Roman(12; 1% of non-emptyNumForm): I, IV, II, III, VI, XI
NumForm seems to be lexical feature of X. 100% lemmas (277) occur only with one value of NumForm.
Relations with Agreement in NumForm
The 10 most frequent relations where parent and child node agree in NumForm:
X –[conj]–> X (100; 100%),
NUM –[conj]–> NUM (75; 100%),
ADJ –[conj]–> ADJ (43; 100%),
NUM –[flat]–> NUM (18; 100%),
X –[flat]–> X (11; 100%),
X –[amod]–> X (6; 100%),
ADJ –[fixed]–> X (1; 100%),
NUM –[nummod:gov]–> NUM (1; 100%),
X –[parataxis:insert]–> X (1; 100%).