home ja/pos edit page issue tracker

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.

Cardinal numerals are convered by NUM, which are tagged with noun(numeral) / 名詞-数詞 in UniDic, including Kanji expressions (e.g. 二十 “20”, 六万 “60,000”).

Note that each numeral is split into one or more word units based on the number of units, (e.g. 二千十四 “2014” is split into three words, _二千 “2000”, 十 “10”　and 四 “4”).

Examples

References

Treebank Statistics (UD_Japanese-KTC)

There are 12 NUM lemmas (0%), 1 NUM types (6%) and 10685 NUM tokens (4%). Out of 16 observed tags, the rank of NUM is: 10 in number of lemmas, 9 in number of types and 6 in number of tokens.

The 10 most frequent NUM lemmas: _, 零, 一, 二, 三, 五, 四, 九, 六, 八

The 10 most frequent NUM types: _

The 10 most frequent ambiguous lemmas: _ (NOUN 73916, ADP 56601, PUNCT 29066, AUX 10360, SCONJ 9060, NUM 8667, VERB 8579, ADJ 3304, PART 2730, CONJ 2110, PROPN 1795, ADV 1655, SYM 1138, PRON 138, DET 95, INTJ 15), 一 (NUM 361, PART 2), 九 (NUM 96, NOUN 1), 八 (NUM 92, PART 1)

The 10 most frequent ambiguous types: _ (NOUN 83879, ADP 56602, AUX 29224, PUNCT 29066, VERB 24527, NUM 10685, SCONJ 9342, PROPN 7729, ADJ 4996, PART 2783, CONJ 2763, ADV 2738, SYM 1138, DET 1067, PRON 1065, INTJ 27)

Morphology

The form / lemma ratio of NUM is 0.083333 (the average of all parts of speech is 0.002927).

The 1st highest number of forms (1) was observed with the lemma “_”: _.

The 2nd highest number of forms (1) was observed with the lemma “ゼロ-zero”: _.

The 3rd highest number of forms (1) was observed with the lemma “一”: _.

NUM does not occur with any features.

Relations

NUM nodes are attached to their parents using 14 different relations: ja-dep/nummod (10391; 97% instances), ja-dep/nmod (109; 1% instances), ja-dep/compound (39; 0% instances), ja-dep/dep (39; 0% instances), ja-dep/nsubj (20; 0% instances), ja-dep/iobj (19; 0% instances), ja-dep/root (18; 0% instances), ja-dep/conj (16; 0% instances), ja-dep/dobj (14; 0% instances), ja-dep/appos (12; 0% instances), ja-dep/ccomp (4; 0% instances), ja-dep/advcl (2; 0% instances), ja-dep/acl (1; 0% instances), ja-dep/advmod (1; 0% instances)

Parents of NUM nodes belong to 7 different parts of speech: NOUN (5415; 51% instances), NUM (5151; 48% instances), VERB (87; 1% instances), ROOT (18; 0% instances), PROPN (8; 0% instances), ADJ (5; 0% instances), PRON (1; 0% instances)

5394 (50%) NUM nodes are leaves.

4783 (45%) NUM nodes have one child.

339 (3%) NUM nodes have two children.

169 (2%) NUM nodes have three or more children.

The highest child degree of a NUM node is 9.

Children of NUM nodes are attached using 18 different relations: ja-dep/nummod (5136; 84% instances), ja-dep/dep (429; 7% instances), ja-dep/punct (285; 5% instances), ja-dep/case (93; 2% instances), ja-dep/compound (81; 1% instances), ja-dep/nsubj (28; 0% instances), ja-dep/conj (25; 0% instances), ja-dep/nmod (23; 0% instances), ja-dep/acl (5; 0% instances), ja-dep/cop (4; 0% instances), ja-dep/appos (3; 0% instances), ja-dep/aux (3; 0% instances), ja-dep/advmod (2; 0% instances), ja-dep/cc (2; 0% instances), ja-dep/det (2; 0% instances), ja-dep/mark (2; 0% instances), ja-dep/csubj (1; 0% instances), ja-dep/iobj (1; 0% instances)

Children of NUM nodes belong to 15 different parts of speech: NUM (5151; 84% instances), NOUN (431; 7% instances), PUNCT (217; 4% instances), ADP (190; 3% instances), SYM (68; 1% instances), PROPN (34; 1% instances), VERB (10; 0% instances), AUX (8; 0% instances), PART (4; 0% instances), ADV (3; 0% instances), CONJ (2; 0% instances), DET (2; 0% instances), PRON (2; 0% instances), SCONJ (2; 0% instances), ADJ (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]