Treebank Statistics: UD_Spanish-AnCora: Features: NumForm
This feature is language-specific.
It occurs with 3 different values: Digit, Roman, Word.
9197 tokens (2%) have a non-empty value of NumForm.
1640 types (4%) occur at least once with a non-empty value of NumForm.
1617 lemmas (6%) occur at least once with a non-empty value of NumForm.
The feature is used with 3 part-of-speech tags: NUM (9173; 2% instances), SYM (20; 0% instances), ADJ (4; 0% instances).
NUM
9173 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: NumType=Card (9171; 100%), Number=EMPTY (6313; 69%).
NUM tokens may have the following values of NumForm:
Digit(5473; 60% of non-emptyNumForm): 20, 30, 10, 15, 12, 2, 25, 50, 16, 18Roman(121; 1% of non-emptyNumForm): II, XX, V, VI, XXI, XVIII, XXIII, IX, XIX, XVWord(3579; 39% of non-emptyNumForm): dos, ciento, tres, cinco, cuatro, seis, ambos, siete, diez, ocho
NumForm seems to be lexical feature of NUM. 100% lemmas (1613) occur only with one value of NumForm.
SYM
20 SYM tokens (5% of all SYM tokens) have a non-empty value of NumForm.
SYM tokens may have the following values of NumForm:
Digit(20; 100% of non-emptyNumForm): %, -0,7%, -0,9%, -1,1%
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: Gender=EMPTY (4; 100%), Number=EMPTY (4; 100%), VerbForm=EMPTY (4; 100%).
ADJ tokens may have the following values of NumForm:
Digit(4; 100% of non-emptyNumForm): 1ro, 2do, 3ro, 4to
Relations with Agreement in NumForm
The 10 most frequent relations where parent and child node agree in NumForm:
NUM –[conj]–> NUM (195; 98%),
NUM –[compound]–> NUM (156; 96%),
NUM –[appos]–> NUM (23; 96%),
NUM –[nmod]–> SYM (17; 100%),
NUM –[nummod]–> NUM (12; 80%),
NUM –[list]–> NUM (10; 100%),
ADJ –[list]–> ADJ (3; 100%),
NUM –[obl]–> NUM (1; 100%).