home edit page issue tracker

This page pertains to UD version 2.

Treebank Statistics: UD_Czech-PDTC: POS Tags: NUM

There are 7095 NUM lemmas (8%), 7234 NUM types (4%) and 104456 NUM tokens (3%). Out of 17 observed tags, the rank of NUM is: 6 in number of lemmas, 6 in number of types and 12 in number of tokens.

The 10 most frequent NUM lemmas: jeden, milion, dva, 1, tři, miliarda, 2, oba, 3, 4

The 10 most frequent NUM types: 1, milionů, milionu, dva, tři, 2, jeden, miliardy, 3, 4

The 10 most frequent ambiguous lemmas: pět (NUM 1420, VERB 3), sto (NUM 415, ADJ 1), I (NOUN 137, NUM 101, X 7, PROPN 2), V (NOUN 296, NUM 19), XX (NUM 4, NOUN 1), XL (NOUN 4, NUM 1)

The 10 most frequent ambiguous types: dvou (NUM 987, ADJ 2), pěti (NUM 502, ADJ 3), tří (NUM 438, ADJ 9), jedno (NUM 293, ADV 29, ADJ 2), šesti (NUM 312, ADJ 2), jednou (ADV 546, NUM 229), deseti (NUM 230, ADJ 1), čtyř (NUM 198, ADJ 3), sto (NUM 183, ADJ 1), sedmi (NUM 177, ADJ 1)

Morphology

The form / lemma ratio of NUM is 1.019591 (the average of all parts of speech is 2.169184).

The 1st highest number of forms (10) was observed with the lemma “jeden”: jeden, jedna, jedno, jednoho, jednom, jednomu, jednou, jednu, jedné, jedním.

The 2nd highest number of forms (10) was observed with the lemma “miliarda”: miliard, miliarda, miliardami, miliardou, miliardu, miliardy, miliardách, miliardám, miliardě, mld.

The 3rd highest number of forms (8) was observed with the lemma “milión”: mil, milión, miliónech, miliónem, miliónu, milióny, miliónů, miliónům.

NUM occurs with 8 features: NumType (104456; 100% instances), NumForm (93184; 89% instances), Case (34484; 33% instances), Number (34484; 33% instances), Gender (21697; 21% instances), Animacy (8736; 8% instances), Abbr (292; 0% instances), Style (4; 0% instances)

NUM occurs with 25 feature-value pairs: Abbr=Yes, Animacy=Anim, Animacy=Inan, Case=Acc, Case=Dat, Case=Gen, Case=Ins, Case=Loc, Case=Nom, Case=Voc, Gender=Fem, Gender=Fem,Neut, Gender=Masc, Gender=Masc,Neut, Gender=Neut, NumForm=Digit, NumForm=Roman, NumForm=Word, NumType=Card, NumType=Frac, NumType=Sets, Number=Dual, Number=Plur, Number=Sing, Style=Coll

NUM occurs with 93 feature combinations. The most frequent feature combination is NumForm=Digit|NumType=Card (67981 tokens). Examples: 1, 2, 3, 4, 8, 10, 30, 5, 15, 20

Relations

NUM nodes are attached to their parents using 25 different relations: nummod (38403; 37% instances), nummod:gov (31918; 31% instances), compound (10800; 10% instances), conj (6990; 7% instances), obl (4896; 5% instances), obl:arg (2014; 2% instances), nsubj (1992; 2% instances), root (1800; 2% instances), obj (1449; 1% instances), dep (1110; 1% instances), parataxis (1009; 1% instances), appos (795; 1% instances), orphan (425; 0% instances), nsubj:pass (189; 0% instances), advcl:pred (182; 0% instances), nmod (155; 0% instances), advcl (130; 0% instances), ccomp (82; 0% instances), acl:relcl (45; 0% instances), xcomp (30; 0% instances), acl (19; 0% instances), iobj (13; 0% instances), csubj (6; 0% instances), csubj:pass (3; 0% instances), vocative (1; 0% instances)

Parents of NUM nodes belong to 16 different parts of speech: NOUN (69824; 67% instances), NUM (14641; 14% instances), VERB (9155; 9% instances), PROPN (2416; 2% instances), ADJ (2344; 2% instances), SYM (2103; 2% instances), (1800; 2% instances), ADV (809; 1% instances), PRON (521; 0% instances), X (506; 0% instances), DET (174; 0% instances), AUX (81; 0% instances), PART (39; 0% instances), CCONJ (36; 0% instances), ADP (6; 0% instances), INTJ (1; 0% instances)

56358 (54%) NUM nodes are leaves.

30449 (29%) NUM nodes have one child.

10625 (10%) NUM nodes have two children.

7024 (7%) NUM nodes have three or more children.

The highest child degree of a NUM node is 57.

Children of NUM nodes are attached using 30 different relations: punct (21244; 27% instances), nmod (15368; 20% instances), compound (8765; 11% instances), case (7938; 10% instances), conj (7367; 9% instances), cc (3518; 4% instances), advmod (3517; 4% instances), advmod:emph (2627; 3% instances), appos (1987; 3% instances), amod (1233; 2% instances), cop (1186; 2% instances), nsubj (1020; 1% instances), dep (800; 1% instances), mark (587; 1% instances), obl (314; 0% instances), det (242; 0% instances), orphan (205; 0% instances), acl:relcl (203; 0% instances), parataxis (116; 0% instances), aux (110; 0% instances), det:nummod (58; 0% instances), csubj (47; 0% instances), advcl (43; 0% instances), acl (13; 0% instances), advcl:pred (7; 0% instances), obl:arg (7; 0% instances), discourse (4; 0% instances), ccomp (3; 0% instances), vocative (3; 0% instances), nummod (2; 0% instances)

Children of NUM nodes belong to 16 different parts of speech: PUNCT (21244; 27% instances), NUM (14641; 19% instances), NOUN (11591; 15% instances), ADP (7940; 10% instances), SYM (5536; 7% instances), PART (4434; 6% instances), ADV (3725; 5% instances), CCONJ (3419; 4% instances), ADJ (1621; 2% instances), AUX (1308; 2% instances), DET (694; 1% instances), PROPN (605; 1% instances), SCONJ (583; 1% instances), VERB (509; 1% instances), X (343; 0% instances), PRON (341; 0% instances)