home edit page issue tracker

This page pertains to UD version 2.

Treebank Statistics: UD_Portuguese-Bosque: POS Tags: NUM

There are 836 NUM lemmas (4%), 849 NUM types (3%) and 4637 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: um, dois, milhão, três, mil, cento, quatro, cinco, 15, 30

The 10 most frequent NUM types: um, dois, milhões, três, mil, uma, cento, duas, quatro, cinco

The 10 most frequent ambiguous lemmas: um (DET 3193, NUM 379, PRON 15, PROPN 3, ADP 2, ADV 1, NOUN 1), dois (NUM 349, NOUN 1), mil (NUM 189, ADP 1), cento (NUM 132, PROPN 3), cinco (NUM 83, PROPN 1), 30 (NUM 59, PROPN 1), 20 (NUM 50, PROPN 3), uma (NUM 47, ADP 12, DET 11, NOUN 4, ADJ 1, PROPN 1, SCONJ 1), 1 (NUM 42, PROPN 13), sete (NUM 37, PROPN 2)

The 10 most frequent ambiguous types: um (DET 1597, NUM 241, PRON 6, PROPN 3, ADP 2, ADV 1, NOUN 1), dois (NUM 222, NOUN 1), mil (NUM 188, ADP 1), uma (DET 1418, NUM 136, ADP 10, NOUN 3, PRON 3, ADJ 1, PROPN 1), cento (NUM 132, PROPN 3), cinco (NUM 77, PROPN 1), 30 (NUM 59, PROPN 1), 20 (NUM 50, PROPN 3), 1 (NUM 41, PROPN 13), sete (NUM 34, PROPN 2)

Morphology

The form / lemma ratio of NUM is 1.015550 (the average of all parts of speech is 1.414227).

The 1st highest number of forms (3) was observed with the lemma “milhão”: mi, milhão, milhões.

The 2nd highest number of forms (2) was observed with the lemma “0”: 0, O.

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

NUM occurs with 3 features: NumType (4623; 100% instances), Number (163; 4% instances), Gender (158; 3% instances)

NUM occurs with 11 feature-value pairs: Gender=Fem, Gender=Masc, Gender=Unsp, NumType=Card, NumType=Frac, NumType=Mult, NumType=Ord, NumType=Range, NumType=Sets, Number=Plur, Number=Sing

NUM occurs with 20 feature combinations. The most frequent feature combination is NumType=Card (4466 tokens). Examples: um, dois, três, milhões, mil, uma, duas, quatro, cinco, 15

Relations

NUM nodes are attached to their parents using 25 different relations: nummod (2629; 57% instances), flat (456; 10% instances), nmod (422; 9% instances), obl (336; 7% instances), conj (138; 3% instances), compound (133; 3% instances), obj (119; 3% instances), nsubj (89; 2% instances), parataxis (89; 2% instances), appos (87; 2% instances), root (57; 1% instances), fixed (24; 1% instances), nsubj:pass (14; 0% instances), xcomp (11; 0% instances), flat:name (10; 0% instances), iobj (6; 0% instances), advcl (3; 0% instances), orphan (3; 0% instances), acl:relcl (2; 0% instances), ccomp (2; 0% instances), list (2; 0% instances), obl:agent (2; 0% instances), acl (1; 0% instances), advmod (1; 0% instances), goeswith (1; 0% instances)

Parents of NUM nodes belong to 13 different parts of speech: NOUN (2447; 53% instances), NUM (643; 14% instances), VERB (535; 12% instances), SYM (424; 9% instances), ADP (272; 6% instances), PROPN (162; 3% instances), (57; 1% instances), ADJ (47; 1% instances), DET (19; 0% instances), PRON (16; 0% instances), ADV (10; 0% instances), X (4; 0% instances), PART (1; 0% instances)

2795 (60%) NUM nodes are leaves.

1052 (23%) NUM nodes have one child.

449 (10%) NUM nodes have two children.

341 (7%) NUM nodes have three or more children.

The highest child degree of a NUM node is 9.

Children of NUM nodes are attached using 28 different relations: case (813; 25% instances), punct (585; 18% instances), nmod (570; 18% instances), flat (459; 14% instances), advmod (218; 7% instances), conj (138; 4% instances), cc (101; 3% instances), det (90; 3% instances), cop (63; 2% instances), nsubj (50; 2% instances), appos (41; 1% instances), acl:relcl (13; 0% instances), amod (12; 0% instances), parataxis (12; 0% instances), advcl (11; 0% instances), mark (9; 0% instances), nummod (9; 0% instances), fixed (6; 0% instances), acl (4; 0% instances), ccomp (4; 0% instances), compound (4; 0% instances), obl (3; 0% instances), xcomp (3; 0% instances), aux (2; 0% instances), flat:name (2; 0% instances), csubj (1; 0% instances), dislocated (1; 0% instances), goeswith (1; 0% instances)

Children of NUM nodes belong to 15 different parts of speech: ADP (822; 25% instances), NUM (643; 20% instances), PUNCT (585; 18% instances), NOUN (534; 17% instances), ADV (216; 7% instances), CCONJ (104; 3% instances), DET (96; 3% instances), AUX (65; 2% instances), PROPN (52; 2% instances), VERB (41; 1% instances), PRON (36; 1% instances), ADJ (14; 0% instances), SCONJ (9; 0% instances), SYM (7; 0% instances), X (1; 0% instances)