home fr/pos edit page issue tracker

NUM: numeral

Definition

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

Cardinal numerals are covered by NUM regardless of syntactic function and regardless of whether they are expressed as words (quatre “four”), digits (4) or Roman numerals (IV). By contrast, ordinal numerals are always tagged ADJ.

Note: Numerals are not yet consistently annotated in the French UD treebank, and sometimes appear as ADJ or DET.


Treebank Statistics (UD_French)

There are 1945 NUM lemmas (5%), 1945 NUM types (4%) and 10791 NUM tokens (3%). Out of 17 observed tags, the rank of NUM is: 5 in number of lemmas, 5 in number of types and 10 in number of tokens.

The 10 most frequent NUM lemmas: deux, trois, 2, 3, 5, 4, quatre, 2010, 2009, 2008

The 10 most frequent NUM types: deux, trois, 2, 3, 5, 4, quatre, 2010, 2009, 2008

The 10 most frequent ambiguous lemmas: deux (NUM 599, PRON 10, NOUN 4), trois (NUM 227, PRON 2), 2 (NUM 160, PROPN 5, PRON 1), 4 (NUM 125, PROPN 2), quatre (NUM 115, PRON 1), 2010 (NUM 122, PROPN 1), 1 (NUM 112, PROPN 2, X 1), 15 (NUM 94, PROPN 1), 12 (NUM 79, PROPN 1), 18 (NUM 78, PROPN 1)

The 10 most frequent ambiguous types: deux (NUM 599, PRON 6, NOUN 4), trois (NUM 227, PRON 2), 2 (NUM 160, PROPN 5, PRON 2), 4 (NUM 125, PROPN 2, ADJ 1), quatre (NUM 115, PRON 1), 2010 (NUM 122, PROPN 1), 1 (NUM 112, PROPN 2, X 1, ADJ 1), 7 (NUM 108, ADJ 1), 10 (NUM 100, ADJ 1), 15 (NUM 94, PROPN 1)

Morphology

The form / lemma ratio of NUM is 1.000000 (the average of all parts of speech is 1.308112).

The 1st highest number of forms (2) was observed with the lemma “1963”: 1963, 1969.

The 2nd highest number of forms (1) was observed with the lemma “’06”: ‘06.

The 3rd highest number of forms (1) was observed with the lemma “’900”: ‘900.

NUM occurs with 2 features: fr-feat/Gender (10; 0% instances), fr-feat/Number (10; 0% instances)

NUM occurs with 2 feature-value pairs: Gender=Fem, Number=Plur

NUM occurs with 2 feature combinations. The most frequent feature combination is _ (10781 tokens). Examples: deux, trois, 2, 3, 5, 4, quatre, 2010, 2009, 2008

Relations

NUM nodes are attached to their parents using 20 different relations: fr-dep/nummod (5497; 51% instances), fr-dep/nmod (4533; 42% instances), fr-dep/conj (387; 4% instances), fr-dep/appos (117; 1% instances), fr-dep/compound (65; 1% instances), fr-dep/dobj (52; 0% instances), fr-dep/nsubj (40; 0% instances), fr-dep/root (38; 0% instances), fr-dep/nsubjpass (17; 0% instances), fr-dep/name (11; 0% instances), fr-dep/dep (10; 0% instances), fr-dep/acl (8; 0% instances), fr-dep/acl:relcl (3; 0% instances), fr-dep/amod (3; 0% instances), fr-dep/xcomp (3; 0% instances), fr-dep/advcl (2; 0% instances), fr-dep/mwe (2; 0% instances), fr-dep/ccomp (1; 0% instances), fr-dep/det (1; 0% instances), fr-dep/parataxis (1; 0% instances)

Parents of NUM nodes belong to 14 different parts of speech: NOUN (6819; 63% instances), VERB (1997; 19% instances), PROPN (741; 7% instances), NUM (639; 6% instances), SYM (390; 4% instances), ADJ (69; 1% instances), X (47; 0% instances), ROOT (38; 0% instances), PRON (33; 0% instances), ADP (7; 0% instances), ADV (5; 0% instances), INTJ (3; 0% instances), PUNCT (2; 0% instances), DET (1; 0% instances)

6841 (63%) NUM nodes are leaves.

2696 (25%) NUM nodes have one child.

892 (8%) NUM nodes have two children.

362 (3%) NUM nodes have three or more children.

The highest child degree of a NUM node is 15.

Children of NUM nodes are attached using 26 different relations: fr-dep/case (2789; 47% instances), fr-dep/punct (1114; 19% instances), fr-dep/nmod (437; 7% instances), fr-dep/conj (392; 7% instances), fr-dep/advmod (378; 6% instances), fr-dep/cc (273; 5% instances), fr-dep/nummod (189; 3% instances), fr-dep/det (122; 2% instances), fr-dep/cop (55; 1% instances), fr-dep/nsubj (42; 1% instances), fr-dep/appos (40; 1% instances), fr-dep/acl (21; 0% instances), fr-dep/amod (17; 0% instances), fr-dep/compound (15; 0% instances), fr-dep/mark (7; 0% instances), fr-dep/dobj (6; 0% instances), fr-dep/acl:relcl (4; 0% instances), fr-dep/aux (4; 0% instances), fr-dep/advcl (3; 0% instances), fr-dep/ccomp (2; 0% instances), fr-dep/expl (2; 0% instances), fr-dep/mwe (2; 0% instances), fr-dep/auxpass (1; 0% instances), fr-dep/nsubjpass (1; 0% instances), fr-dep/parataxis (1; 0% instances), fr-dep/xcomp (1; 0% instances)

Children of NUM nodes belong to 17 different parts of speech: ADP (2801; 47% instances), PUNCT (1114; 19% instances), NUM (639; 11% instances), ADV (367; 6% instances), NOUN (324; 5% instances), CONJ (266; 4% instances), DET (127; 2% instances), VERB (93; 2% instances), PROPN (63; 1% instances), PRON (51; 1% instances), ADJ (25; 0% instances), X (23; 0% instances), SYM (11; 0% instances), SCONJ (6; 0% instances), AUX (5; 0% instances), PART (2; 0% instances), INTJ (1; 0% instances)


NUM in other languages: [bg] [cs] [de] [el] [en] [es] [eu] [fa] [fi] [fr] [ga] [he] [hu] [it] [ja] [ko] [sv] [u]