home edit page issue tracker

This page pertains to UD version 2.

Treebank Statistics: UD_Portuguese-PetroGold: POS Tags: NUM

There are 1342 NUM lemmas (11%), 1346 NUM types (8%) and 7290 NUM tokens (3%). Out of 16 observed tags, the rank of NUM is: 4 in number of lemmas, 5 in number of types and 8 in number of tokens.

The 10 most frequent NUM lemmas: dois, 1, 3, 2, 5, 10, 4, três, 2005, 20

The 10 most frequent NUM types: 1, dois, 3, 2, 5, 10, duas, 4, três, 2005

The 10 most frequent ambiguous lemmas: um (DET 2760, PRON 120, NUM 49), milhão (NUM 40, NOUN 1), p/p (NUM 31, NOUN 13), 3.3 (NUM 22, PROPN 1), 2D (NUM 9, NOUN 2), 6.5 (NUM 7, PROPN 2), III.3 (PROPN 7, NUM 6), I (PROPN 29, NUM 5), 2.000 (NUM 4, PROPN 1), V.2 (NUM 4, PROPN 1)

The 10 most frequent ambiguous types: milhões (NUM 32, NOUN 1), p/p (NUM 31, NOUN 13), um (DET 1351, PRON 54, NUM 27), 3.3 (NUM 22, PROPN 1), uma (DET 1295, PRON 47, NUM 21), 3D (PROPN 24, NUM 11, NOUN 6, ADJ 1), 2D (NUM 9, NOUN 6), 6.5 (NUM 7, PROPN 2), III.3 (PROPN 7, NUM 6), I (PROPN 29, NUM 5)

Morphology

The form / lemma ratio of NUM is 1.002981 (the average of all parts of speech is 1.452143).

The 1st highest number of forms (2) was observed with the lemma “dois”: dois, duas.

The 2nd highest number of forms (2) was observed with the lemma “milhão”: milhão, milhões.

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

NUM occurs with 3 features: NumType (7245; 99% instances), Number (56; 1% instances), Gender (51; 1% instances)

NUM occurs with 6 feature-value pairs: Gender=Fem, Gender=Masc, NumType=Card, NumType=Ord, NumType=Range, Number=Sing

NUM occurs with 10 feature combinations. The most frequent feature combination is NumType=Card (7207 tokens). Examples: dois, 1, 3, 2, 5, 10, duas, três, 4, 2005

Relations

NUM nodes are attached to their parents using 16 different relations: nummod (4766; 65% instances), nmod (1200; 16% instances), conj (493; 7% instances), flat:name (284; 4% instances), appos (149; 2% instances), obl (140; 2% instances), root (81; 1% instances), flat (66; 1% instances), parataxis (47; 1% instances), nsubj (18; 0% instances), obj (14; 0% instances), obl:arg (13; 0% instances), advcl (7; 0% instances), nsubj:pass (7; 0% instances), xcomp (4; 0% instances), amod (1; 0% instances)

Parents of NUM nodes belong to 11 different parts of speech: NOUN (4042; 55% instances), PROPN (1806; 25% instances), SYM (573; 8% instances), NUM (534; 7% instances), VERB (199; 3% instances), (81; 1% instances), ADJ (37; 1% instances), ADV (8; 0% instances), PRON (7; 0% instances), ADP (2; 0% instances), X (1; 0% instances)

4152 (57%) NUM nodes are leaves.

2170 (30%) NUM nodes have one child.

770 (11%) NUM nodes have two children.

198 (3%) NUM nodes have three or more children.

The highest child degree of a NUM node is 11.

Children of NUM nodes are attached using 21 different relations: punct (2537; 57% instances), conj (508; 11% instances), case (466; 10% instances), cc (351; 8% instances), nmod (187; 4% instances), advmod (148; 3% instances), flat (68; 2% instances), obl (34; 1% instances), cop (33; 1% instances), appos (31; 1% instances), nsubj (26; 1% instances), parataxis (15; 0% instances), det (12; 0% instances), mark (11; 0% instances), flat:name (6; 0% instances), nummod (6; 0% instances), acl:relcl (2; 0% instances), advcl (2; 0% instances), amod (2; 0% instances), orphan (2; 0% instances), acl (1; 0% instances)

Children of NUM nodes belong to 14 different parts of speech: PUNCT (2537; 57% instances), NUM (534; 12% instances), ADP (503; 11% instances), CCONJ (347; 8% instances), NOUN (230; 5% instances), ADV (153; 3% instances), PROPN (34; 1% instances), AUX (33; 1% instances), DET (19; 0% instances), VERB (18; 0% instances), SYM (13; 0% instances), ADJ (10; 0% instances), SCONJ (9; 0% instances), PRON (8; 0% instances)