home edit page issue tracker

This page pertains to UD version 2.

Treebank Statistics: UD_Spanish-AnCora: POS Tags: NUM

There are 1603 NUM lemmas (6%), 1601 NUM types (4%) and 9174 NUM tokens (2%). Out of 17 observed tags, the rank of NUM is: 5 in number of lemmas, 5 in number of types and 13 in number of tokens.

The 10 most frequent NUM lemmas: dos, ciento, tres, cinco, cuatro, ambos, seis, 20, 30, siete

The 10 most frequent NUM types: dos, ciento, tres, cinco, cuatro, seis, 20, 30, ambos, siete

The 10 most frequent ambiguous lemmas: dos (NUM 951, DET 1), tres (NUM 454, NOUN 1), cinco (NUM 236, NOUN 3), cuatro (NUM 228, NOUN 1), seis (NUM 157, NOUN 1), 30 (NUM 114, NOUN 1), siete (NUM 107, NOUN 1), 10 (NUM 99, NOUN 2), medio (NOUN 170, NUM 94, ADJ 38, ADV 7, PROPN 1), 12 (NUM 89, NOUN 1, SYM 1)

The 10 most frequent ambiguous types: dos (NUM 927, NOUN 1), tres (NUM 432, NOUN 1), cinco (NUM 231, NOUN 3), cuatro (NUM 220, NOUN 1), seis (NUM 151, NOUN 1), 30 (NUM 114, NOUN 1), siete (NUM 102, NOUN 1), 10 (NUM 99, NOUN 2), 12 (NUM 87, NOUN 1), 2 (NUM 76, NOUN 1)

Morphology

The form / lemma ratio of NUM is 0.998752 (the average of all parts of speech is 1.505808).

The 1st highest number of forms (3) was observed with the lemma “13”: 13, 13-J, 13`.

The 2nd highest number of forms (3) was observed with the lemma “4”: 4, 4A, 4B.

The 3rd highest number of forms (2) was observed with the lemma “-1”: 3-3-3-1, 4-2-3-1.

NUM occurs with 5 features: NumType (9174; 100% instances), NumForm (9173; 100% instances), Number (2861; 31% instances), Gender (290; 3% instances), Foreign (2; 0% instances)

NUM occurs with 10 feature-value pairs: Foreign=Yes, Gender=Fem, Gender=Masc, NumForm=Digit, NumForm=Roman, NumForm=Word, NumType=Card, NumType=Frac, Number=Plur, Number=Sing

NUM occurs with 14 feature combinations. The most frequent feature combination is NumForm=Digit|NumType=Card (5468 tokens). Examples: 20, 30, 10, 15, 12, 2, 25, 50, 16, 18

Relations

NUM nodes are attached to their parents using 21 different relations: nummod (6000; 65% instances), obl (750; 8% instances), nmod (551; 6% instances), compound (471; 5% instances), appos (389; 4% instances), conj (317; 3% instances), obj (237; 3% instances), nsubj (231; 3% instances), obl:arg (66; 1% instances), flat (53; 1% instances), root (34; 0% instances), dep (23; 0% instances), list (10; 0% instances), fixed (8; 0% instances), advcl (7; 0% instances), obl:agent (7; 0% instances), acl (6; 0% instances), parataxis (5; 0% instances), ccomp (4; 0% instances), xcomp (4; 0% instances), csubj (1; 0% instances)

Parents of NUM nodes belong to 14 different parts of speech: NOUN (5228; 57% instances), VERB (1170; 13% instances), NUM (880; 10% instances), DET (749; 8% instances), SYM (355; 4% instances), PROPN (342; 4% instances), ADV (204; 2% instances), ADJ (166; 2% instances), (34; 0% instances), PRON (26; 0% instances), ADP (11; 0% instances), AUX (6; 0% instances), CCONJ (2; 0% instances), X (1; 0% instances)

5407 (59%) NUM nodes are leaves.

1795 (20%) NUM nodes have one child.

771 (8%) NUM nodes have two children.

1201 (13%) NUM nodes have three or more children.

The highest child degree of a NUM node is 10.

Children of NUM nodes are attached using 24 different relations: case (1822; 23% instances), nmod (1725; 22% instances), det (1626; 21% instances), punct (987; 13% instances), cc (284; 4% instances), conj (263; 3% instances), compound (248; 3% instances), amod (220; 3% instances), advmod (166; 2% instances), appos (126; 2% instances), flat (68; 1% instances), advcl (67; 1% instances), cop (45; 1% instances), obl (37; 0% instances), nsubj (35; 0% instances), acl (18; 0% instances), mark (17; 0% instances), nummod (15; 0% instances), list (10; 0% instances), parataxis (4; 0% instances), aux (2; 0% instances), csubj (1; 0% instances), fixed (1; 0% instances), obj (1; 0% instances)

Children of NUM nodes belong to 14 different parts of speech: ADP (1837; 24% instances), DET (1635; 21% instances), NOUN (1088; 14% instances), PUNCT (987; 13% instances), NUM (880; 11% instances), PRON (320; 4% instances), CCONJ (292; 4% instances), ADJ (252; 3% instances), PROPN (174; 2% instances), ADV (147; 2% instances), VERB (87; 1% instances), AUX (48; 1% instances), SYM (26; 0% instances), SCONJ (15; 0% instances)