home cs/pos edit page issue tracker

This page pertains to UD version 2.

NUM: numeral

Definition

A numeral is a word, functioning most typically as a determiner, adjective or pronoun, that expresses a number and a relation to the number, such as quantity, sequence, frequency or fraction.

Note that cardinal numerals are covered by NUM whether they are used as determiners or not (as in Windows 7) and whether they are expressed as words (čtyři), digits (4) or Roman numerals (IV).

Czech grammar distinguishes several subclasses of pronominal numerals (quantifiers): interrogative and relative (kolik  “how many”); demonstrative (tolik  “this many”); indefinite (několik, mnoho, málo  “several, many, few”). These words behave similarly to (most) cardinal numbers, e.g. they require that the counted noun phrase be in genitive. They are not similar to adjectives (unlike their English counterparts). However, in accord with the UD standard, they should be tagged DET, not NUM.

In addition, several types of (non-pronominal) numerals, such as ordinal numerals and multiplicative numerals, are tagged ADJ or ADV, based on their syntactic and morphological behavior.

Examples

Counterexamples

References


Treebank Statistics (UD_Czech)

There are 3253 NUM lemmas (6%), 3358 NUM types (3%) and 36842 NUM tokens (3%). Out of 17 observed tags, the rank of NUM is: 5 in number of lemmas, 5 in number of types and 11 in number of tokens.

The 10 most frequent NUM lemmas: jeden, dva, 1, tři, 2, oba, 3, 4, pět, 6

The 10 most frequent NUM types: 1, 2, 3, dva, tři, 4, jeden, 6, dvě, 5

The 10 most frequent ambiguous lemmas: jeden (NUM 2229, ADJ 28), pět (NUM 560, VERB 1), tisíc (NUM 465, NOUN 284, ADV 1), osm (NUM 203, ADJ 1), I (NUM 90, PROPN 59, ADJ 16, PRON 14), půl (NOUN 156, NUM 53), třináct (NUM 50, ADJ 1), sto (NOUN 275, NUM 38), šestnáct (NUM 25, ADJ 1), V (PROPN 183, NUM 21, NOUN 7, ADJ 5)

The 10 most frequent ambiguous types: tisíc (NUM 464, NOUN 76), dvou (NUM 457, ADJ 1), tří (NUM 223, ADJ 2), jedno (NUM 133, ADJ 1), jednou (ADV 145, NUM 117), I (CCONJ 413, NUM 90, PROPN 59, ADJ 16, PRON 5, NOUN 1), čtyř (NUM 89, ADJ 1), osmi (NUM 81, ADJ 1), půl (NOUN 143, NUM 53), sto (NOUN 107, NUM 23)

Morphology

The form / lemma ratio of NUM is 1.032278 (the average of all parts of speech is 2.162583).

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 (8) was observed with the lemma “třetina”: třetin, třetina, třetinou, třetinu, třetiny, třetinách, třetinám, třetině.

The 3rd highest number of forms (6) was observed with the lemma “sto”: set, stech, sto, sty, stě, stům.

NUM occurs with 10 features: cs-feat/NumType (36842; 100% instances), cs-feat/NumForm (36547; 99% instances), cs-feat/Number (10268; 28% instances), cs-feat/Case (10244; 28% instances), cs-feat/NumValue (7079; 19% instances), cs-feat/Gender (4177; 11% instances), cs-feat/Animacy (278; 1% instances), cs-feat/Foreign (26; 0% instances), cs-feat/NameType (19; 0% instances), cs-feat/Style (2; 0% instances)

NUM occurs with 25 feature-value pairs: Animacy=Anim, Animacy=Inan, Case=Acc, Case=Dat, Case=Gen, Case=Ins, Case=Loc, Case=Nom, Foreign=Yes, Gender=Fem, Gender=Fem,Neut, Gender=Masc, Gender=Masc,Neut, Gender=Neut, NameType=Com, NumForm=Digit, NumForm=Roman, NumForm=Word, NumType=Card, NumType=Frac, NumValue=1,2,3, Number=Dual, Number=Plur, Number=Sing, Style=Arch

NUM occurs with 57 feature combinations. The most frequent feature combination is NumForm=Digit|NumType=Card (26226 tokens). Examples: 1, 2, 3, 4, 6, 5, 10, 1992, 1994, 1993

Relations

NUM nodes are attached to their parents using 23 different relations: cs-dep/nummod (17496; 47% instances), cs-dep/nummod:gov (6518; 18% instances), cs-dep/conj (3363; 9% instances), cs-dep/compound (2589; 7% instances), cs-dep/dep (1699; 5% instances), cs-dep/obl (1656; 4% instances), cs-dep/root (991; 3% instances), cs-dep/obj (785; 2% instances), cs-dep/nsubj (622; 2% instances), cs-dep/orphan (436; 1% instances), cs-dep/appos (272; 1% instances), cs-dep/nmod (113; 0% instances), cs-dep/nsubj:pass (78; 0% instances), cs-dep/xcomp (71; 0% instances), cs-dep/iobj (55; 0% instances), cs-dep/advcl (34; 0% instances), cs-dep/acl (26; 0% instances), cs-dep/ccomp (22; 0% instances), cs-dep/parataxis (8; 0% instances), cs-dep/advmod:emph (4; 0% instances), cs-dep/csubj (2; 0% instances), cs-dep/csubj:pass (1; 0% instances), cs-dep/mark (1; 0% instances)

Parents of NUM nodes belong to 15 different parts of speech: NOUN (23252; 63% instances), NUM (5701; 15% instances), VERB (2926; 8% instances), PROPN (2347; 6% instances), ROOT (991; 3% instances), ADJ (921; 2% instances), ADV (308; 1% instances), SYM (190; 1% instances), PRON (108; 0% instances), DET (56; 0% instances), CCONJ (26; 0% instances), PUNCT (12; 0% instances), ADP (2; 0% instances), INTJ (1; 0% instances), PART (1; 0% instances)

19147 (52%) NUM nodes are leaves.

11826 (32%) NUM nodes have one child.

3333 (9%) NUM nodes have two children.

2536 (7%) NUM nodes have three or more children.

The highest child degree of a NUM node is 27.

Children of NUM nodes are attached using 30 different relations: cs-dep/punct (10498; 37% instances), cs-dep/nmod (3588; 13% instances), cs-dep/conj (3298; 12% instances), cs-dep/compound (2589; 9% instances), cs-dep/case (1859; 7% instances), cs-dep/advmod:emph (1817; 6% instances), cs-dep/cc (1079; 4% instances), cs-dep/dep (612; 2% instances), cs-dep/amod (570; 2% instances), cs-dep/cop (419; 1% instances), cs-dep/nsubj (348; 1% instances), cs-dep/advmod (278; 1% instances), cs-dep/mark (268; 1% instances), cs-dep/orphan (226; 1% instances), cs-dep/appos (209; 1% instances), cs-dep/obl (131; 0% instances), cs-dep/nummod (88; 0% instances), cs-dep/det (64; 0% instances), cs-dep/parataxis (39; 0% instances), cs-dep/acl (32; 0% instances), cs-dep/csubj (28; 0% instances), cs-dep/xcomp (26; 0% instances), cs-dep/det:nummod (21; 0% instances), cs-dep/obj (19; 0% instances), cs-dep/advcl (18; 0% instances), cs-dep/aux (9; 0% instances), cs-dep/discourse (4; 0% instances), cs-dep/fixed (1; 0% instances), cs-dep/flat:foreign (1; 0% instances), cs-dep/vocative (1; 0% instances)

Children of NUM nodes belong to 16 different parts of speech: PUNCT (10498; 37% instances), NUM (5701; 20% instances), NOUN (3813; 14% instances), ADP (1845; 7% instances), ADV (1337; 5% instances), CCONJ (1024; 4% instances), PART (828; 3% instances), SYM (794; 3% instances), ADJ (740; 3% instances), AUX (428; 2% instances), PROPN (342; 1% instances), SCONJ (263; 1% instances), DET (188; 1% instances), VERB (172; 1% instances), PRON (166; 1% instances), INTJ (1; 0% instances)


Treebank Statistics (UD_Czech-CAC)

There are 59 NUM lemmas (0%), 123 NUM types (0%) and 7204 NUM tokens (1%). Out of 16 observed tags, the rank of NUM is: 7 in number of lemmas, 8 in number of types and 13 in number of tokens.

The 10 most frequent NUM lemmas: &camount;, &cyear;, jeden, dva, &clabel;, &cdate;, &cother;, oba, tři, čtyři

The 10 most frequent NUM types: #, dvou, jeden, dvě, tři, dva, obou, jednoho, jedné, jedním

The 10 most frequent ambiguous lemmas: jeden (NUM 741, ADJ 10), tisíc (NUM 48, NOUN 38), půl (NUM 36, NOUN 1), pár (NUM 23, NOUN 12), sto (NOUN 36, NUM 12)

The 10 most frequent ambiguous types: tisíc (NUM 48, NOUN 8), jednou (ADV 52, NUM 23), půl (NUM 36, NOUN 1), pár (NUM 23, NOUN 5), sto (NOUN 8, NUM 5), set (NOUN 19, NUM 4)

Morphology

The form / lemma ratio of NUM is 2.084746 (the average of all parts of speech is 2.180683).

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 (7) was observed with the lemma “třetina”: třetin, třetina, třetinami, třetinou, třetinu, třetiny, třetině.

The 3rd highest number of forms (5) was observed with the lemma “tři”: třech, třem, třemi, tři, tří.

NUM occurs with 7 features: cs-feat/NumType (7204; 100% instances), cs-feat/NumForm (7149; 99% instances), cs-feat/Case (2420; 34% instances), cs-feat/Number (2420; 34% instances), cs-feat/NumValue (1930; 27% instances), cs-feat/Gender (1173; 16% instances), cs-feat/Animacy (98; 1% instances)

NUM occurs with 21 feature-value pairs: Animacy=Anim, Animacy=Inan, Case=Acc, Case=Dat, Case=Gen, Case=Ins, Case=Loc, Case=Nom, Gender=Fem, Gender=Fem,Neut, Gender=Masc, Gender=Masc,Neut, Gender=Neut, NumForm=Digit, NumForm=Word, NumType=Card, NumType=Frac, NumValue=1,2,3, Number=Dual, Number=Plur, Number=Sing

NUM occurs with 49 feature combinations. The most frequent feature combination is NumForm=Digit|NumType=Card (4784 tokens). Examples: #

Relations

NUM nodes are attached to their parents using 24 different relations: cs-dep/nummod (4388; 61% instances), cs-dep/nummod:gov (1164; 16% instances), cs-dep/obl (430; 6% instances), cs-dep/conj (395; 5% instances), cs-dep/obj (162; 2% instances), cs-dep/dep (139; 2% instances), cs-dep/nsubj (113; 2% instances), cs-dep/root (113; 2% instances), cs-dep/orphan (102; 1% instances), cs-dep/nsubj:pass (43; 1% instances), cs-dep/compound (42; 1% instances), cs-dep/advcl (29; 0% instances), cs-dep/xcomp (23; 0% instances), cs-dep/appos (21; 0% instances), cs-dep/iobj (14; 0% instances), cs-dep/acl (8; 0% instances), cs-dep/ccomp (5; 0% instances), cs-dep/parataxis (5; 0% instances), cs-dep/aux (2; 0% instances), cs-dep/nmod (2; 0% instances), cs-dep/aux:pass (1; 0% instances), cs-dep/case (1; 0% instances), cs-dep/cop (1; 0% instances), cs-dep/csubj:pass (1; 0% instances)

Parents of NUM nodes belong to 12 different parts of speech: NOUN (5408; 75% instances), VERB (651; 9% instances), NUM (393; 5% instances), ADJ (245; 3% instances), SYM (222; 3% instances), ROOT (113; 2% instances), ADV (60; 1% instances), PROPN (55; 1% instances), PRON (33; 0% instances), DET (19; 0% instances), PART (4; 0% instances), CCONJ (1; 0% instances)

4575 (64%) NUM nodes are leaves.

1644 (23%) NUM nodes have one child.

644 (9%) NUM nodes have two children.

341 (5%) NUM nodes have three or more children.

The highest child degree of a NUM node is 14.

Children of NUM nodes are attached using 27 different relations: cs-dep/nmod (1319; 31% instances), cs-dep/case (570; 13% instances), cs-dep/advmod:emph (550; 13% instances), cs-dep/conj (435; 10% instances), cs-dep/punct (317; 7% instances), cs-dep/cc (311; 7% instances), cs-dep/cop (157; 4% instances), cs-dep/nsubj (135; 3% instances), cs-dep/amod (103; 2% instances), cs-dep/mark (87; 2% instances), cs-dep/advmod (73; 2% instances), cs-dep/orphan (51; 1% instances), cs-dep/compound (42; 1% instances), cs-dep/obl (37; 1% instances), cs-dep/appos (32; 1% instances), cs-dep/nummod (23; 1% instances), cs-dep/dep (22; 1% instances), cs-dep/det (14; 0% instances), cs-dep/acl (8; 0% instances), cs-dep/xcomp (6; 0% instances), cs-dep/csubj (5; 0% instances), cs-dep/det:nummod (5; 0% instances), cs-dep/advcl (4; 0% instances), cs-dep/parataxis (4; 0% instances), cs-dep/aux (2; 0% instances), cs-dep/discourse (1; 0% instances), cs-dep/obj (1; 0% instances)

Children of NUM nodes belong to 15 different parts of speech: SYM (952; 22% instances), NOUN (580; 13% instances), ADP (569; 13% instances), ADV (396; 9% instances), NUM (393; 9% instances), PART (346; 8% instances), PUNCT (317; 7% instances), CCONJ (224; 5% instances), AUX (158; 4% instances), ADJ (133; 3% instances), SCONJ (85; 2% instances), DET (59; 1% instances), VERB (47; 1% instances), PRON (38; 1% instances), PROPN (17; 0% instances)


Treebank Statistics (UD_Czech-CLTT)

There are 76 NUM lemmas (3%), 86 NUM types (2%) and 310 NUM tokens (1%). Out of 15 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: 1, 3, 2, jeden, 4, 5, 41, 2004, 2008, 31

The 10 most frequent NUM types: 1, 3, 2, jeden, 4, 5, 41, 2004, 2008, 31

The 10 most frequent ambiguous lemmas: 1 (NUM 65, X 45), 3 (X 38, NUM 31), 2 (X 49, NUM 28), 4 (X 21, NUM 12), 5 (X 15, NUM 11), 41 (NUM 7, X 2), 31 (NUM 5, X 1), 60 (NUM 4, X 1), 26 (X 4, NUM 3), 8 (X 12, NUM 3)

The 10 most frequent ambiguous types: 1 (NUM 65, X 45), 3 (X 38, NUM 31), 2 (X 49, NUM 28), 4 (X 21, NUM 12), 5 (X 15, NUM 11), 41 (NUM 7, X 2), 31 (NUM 5, X 1), 60 (NUM 4, X 1), 26 (X 4, NUM 3), 8 (X 12, NUM 3)

Morphology

The form / lemma ratio of NUM is 1.131579 (the average of all parts of speech is 1.685169).

The 1st highest number of forms (7) was observed with the lemma “jeden”: jeden, jedno, jednoho, jednom, jednou, jedné, jedním.

The 2nd highest number of forms (3) was observed with the lemma “dva”: dvou, dvě, dvěma.

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

NUM occurs with 7 features: cs-feat/NumForm (310; 100% instances), cs-feat/NumType (310; 100% instances), cs-feat/Case (46; 15% instances), cs-feat/Number (46; 15% instances), cs-feat/NumValue (38; 12% instances), cs-feat/Gender (32; 10% instances), cs-feat/Animacy (12; 4% instances)

NUM occurs with 17 feature-value pairs: Animacy=Inan, Case=Acc, Case=Gen, Case=Ins, Case=Loc, Case=Nom, Gender=Fem, Gender=Fem,Neut, Gender=Masc, Gender=Masc,Neut, Gender=Neut, NumForm=Roman, NumForm=Word, NumType=Card, NumValue=1,2,3, Number=Plur, Number=Sing

NUM occurs with 14 feature combinations. The most frequent feature combination is NumForm=Roman|NumType=Card (264 tokens). Examples: 1, 3, 2, 4, 5, 41, 2004, 2008, 31, 2005

Relations

NUM nodes are attached to their parents using 10 different relations: cs-dep/nummod (195; 63% instances), cs-dep/conj (30; 10% instances), cs-dep/obl (23; 7% instances), cs-dep/nmod (22; 7% instances), cs-dep/nummod:gov (19; 6% instances), cs-dep/advcl (10; 3% instances), cs-dep/obj (6; 2% instances), cs-dep/advmod (3; 1% instances), cs-dep/nsubj (1; 0% instances), cs-dep/orphan (1; 0% instances)

Parents of NUM nodes belong to 6 different parts of speech: NOUN (237; 76% instances), VERB (32; 10% instances), NUM (29; 9% instances), ADV (6; 2% instances), ADJ (4; 1% instances), X (2; 1% instances)

161 (52%) NUM nodes are leaves.

124 (40%) NUM nodes have one child.

23 (7%) NUM nodes have two children.

2 (1%) NUM nodes have three or more children.

The highest child degree of a NUM node is 7.

Children of NUM nodes are attached using 11 different relations: cs-dep/punct (76; 42% instances), cs-dep/conj (31; 17% instances), cs-dep/cc (24; 13% instances), cs-dep/nmod (22; 12% instances), cs-dep/advmod:emph (12; 7% instances), cs-dep/mark (10; 6% instances), cs-dep/case (2; 1% instances), cs-dep/cop (1; 1% instances), cs-dep/dep (1; 1% instances), cs-dep/nsubj (1; 1% instances), cs-dep/obl (1; 1% instances)

Children of NUM nodes belong to 12 different parts of speech: PUNCT (76; 42% instances), NUM (29; 16% instances), CCONJ (18; 10% instances), NOUN (15; 8% instances), ADV (13; 7% instances), SCONJ (13; 7% instances), SYM (8; 4% instances), PART (3; 2% instances), ADP (2; 1% instances), X (2; 1% instances), AUX (1; 1% instances), PRON (1; 1% instances)


NUM in other languages: [am] [ar] [bg] [bxr] [ca] [ckb] [cop] [cs] [cu] [da] [de] [el] [en] [es] [et] [eu] [fa] [fi] [fo] [fr] [ga] [gl] [got] [grc] [he] [hi] [hr] [hu] [id] [it] [ja] [kk] [kmr] [ko] [la] [lv] [mr] [nl] [no] [pl] [pt] [ro] [ru] [sa] [sk] [sla] [sl] [so] [sr] [sv] [swl] [ta] [tr] [ug] [uk] [u] [urj] [ur] [vi] [yue] [zh]