This is part of archived UD v1 documentation. See http://universaldependencies.org/ for the current version.

home sl/pos 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.

The NUM tag is used for digit numerals (3 / 3.), roman numerals (III / III.), cardinal word numerals (tri “three”) and generic numerals (troje “three sets of”).

Other word types expressing a number or a relation to a number in Slovenian are marked as adjectives (tretji “the third”, trojen “triple”, trikraten “threefold”), adverbs (trikrat “three times”, tretjič “the third time”) or nouns (tretjina “a third”, trojica “triplet”, trojka “number three”).

Examples

• 1, 2, 3
• 1., 2., 3.
• I, II, III
• I., II., III.
• en “one”, dva “two”, tri “three”
• enoje “one set of”, dvoje “two sets of”, troje “three sets of”

Conversion from JOS

The following numerals are converted to NUM: numerals with Form=digit; numerals with Form=roman; numerals with Form=letter and Type=cardinal; numeral with Form=letter, Type=pronominal and lemma en or eden; and numerals with Form=letter, Type=special and lemma not ending in -en.

---

Treebank Statistics (UD_Slovenian)

There are 485 NUM lemmas (3%), 524 NUM types (2%) and 1927 NUM tokens (1%). Out of 16 observed tags, the rank of NUM is: 6 in number of lemmas, 6 in number of types and 14 in number of tokens.

The 10 most frequent NUM lemmas: en, dva, trije, štirje, pet, deset, eden, tisoč, šest, sto

The 10 most frequent NUM types: eno, tri, dveh, dva, ena, eden, tisoč, štiri, štirih, dve

The 10 most frequent ambiguous lemmas: V (NOUN 2, NUM 1), V. (X 3, NUM 1), X (NOUN 1, NUM 1)

The 10 most frequent ambiguous types: V (ADP 423, NOUN 2, NUM 1), V. (X 3, NUM 1), X (NUM 1, NOUN 1), dvajsetih (NUM 1, ADJ 1)

• V
  • ADP 423: V njih iščejo dobro in nemalokrat odkrijejo tudi skrito nadarjenost .
  • NOUN 2: Napetost tega člena je 3,20 V .
  • NUM 1: Nivo preverjanja V daje bolj strmo OC krivuljo in zato zmanjša tveganje kupca .
• V.
  • X 3: M. V.
  • NUM 1: Nikolaj V. je , ko je nastopil papeško službo , plačal več kot 1000 funtov pesniku Francescu Filelfu , da je napisal knjigo zgodb , ki so jih opisali kot “ najbolj nagnusna dela , ki jih je kdaj ustvarila surova kljubovalnost in umazana domišljija “ .
• X
  • NUM 1: Morfologija : Ima majhen , krepek , sploščen trup , nekoliko daljšo spodnjo čeljust , hrbtno plavut z X , 11 plavutnicami in predrepno s III , 7 plavutnicami .
  • NOUN 1: A Za podaljšanje meta od točke X do točke Y sprostimo toliko vrvice ( 1 ) , da z njo dosežemo cilj .
• dvajsetih
  • NUM 1: Pri dvajsetih letih se le redko katera ženska sprašuje o svoji plodnosti , večina se zanositvi prav v teh letih skuša čim bolj izogniti .
  • ADJ 1: Bluesovska zgodovina je v eksploziji “ umazanega ” , “ globokega ” , silovitega bluesa Muddyja Watersa in sodobnikov ob koncu štiridesetih let razbrala ponovni vznik stila , ki ga poznamo iz dvajsetih let , v downhome bluesovski tradiciji .

Morphology

The form / lemma ratio of NUM is 1.080412 (the average of all parts of speech is 1.894262).

The 1st highest number of forms (10) was observed with the lemma “en”: en, ena, ene, enega, enem, enemu, eni, enih, enim, eno.

The 2nd highest number of forms (5) was observed with the lemma “trije”: treh, trem, tremi, tri, trije.

The 3rd highest number of forms (5) was observed with the lemma “štirje”: štiri, štirih, štirim, štirimi, štirje.

NUM occurs with 5 features: sl-feat/NumForm (1927; 100% instances), sl-feat/NumType (1927; 100% instances), sl-feat/Case (740; 38% instances), sl-feat/Number (740; 38% instances), sl-feat/Gender (486; 25% instances)

NUM occurs with 18 feature-value pairs: Case=Acc, Case=Dat, Case=Gen, Case=Ins, Case=Loc, Case=Nom, Gender=Fem, Gender=Masc, Gender=Neut, NumForm=Digit, NumForm=Roman, NumForm=Word, NumType=Card, NumType=Ord, NumType=Sets, Number=Dual, Number=Plur, Number=Sing

NUM occurs with 61 feature combinations. The most frequent feature combination is NumForm=Digit|NumType=Card (921 tokens). Examples: 10, 15, 2000, 50, 30, 3, 20, 6, 40, 2

Relations

NUM nodes are attached to their parents using 12 different relations: sl-dep/nummod (1586; 82% instances), sl-dep/compound (114; 6% instances), sl-dep/conj (101; 5% instances), sl-dep/advmod (66; 3% instances), sl-dep/parataxis (24; 1% instances), sl-dep/nsubj (13; 1% instances), sl-dep/dobj (9; 0% instances), sl-dep/dep (8; 0% instances), sl-dep/root (3; 0% instances), sl-dep/acl (1; 0% instances), sl-dep/iobj (1; 0% instances), sl-dep/nmod (1; 0% instances)

Parents of NUM nodes belong to 9 different parts of speech: NOUN (1558; 81% instances), NUM (128; 7% instances), VERB (109; 6% instances), ADJ (68; 4% instances), PROPN (47; 2% instances), PRON (8; 0% instances), ADV (5; 0% instances), ROOT (3; 0% instances), X (1; 0% instances)

1471 (76%) NUM nodes are leaves.

319 (17%) NUM nodes have one child.

107 (6%) NUM nodes have two children.

30 (2%) NUM nodes have three or more children.

The highest child degree of a NUM node is 7.

Children of NUM nodes are attached using 14 different relations: sl-dep/punct (178; 28% instances), sl-dep/case (124; 19% instances), sl-dep/advmod (116; 18% instances), sl-dep/conj (108; 17% instances), sl-dep/cc (38; 6% instances), sl-dep/compound (34; 5% instances), sl-dep/nmod (18; 3% instances), sl-dep/acl (10; 2% instances), sl-dep/amod (6; 1% instances), sl-dep/cop (2; 0% instances), sl-dep/advcl (1; 0% instances), sl-dep/aux (1; 0% instances), sl-dep/mark (1; 0% instances), sl-dep/nsubj (1; 0% instances)

Children of NUM nodes belong to 14 different parts of speech: PUNCT (178; 28% instances), NUM (128; 20% instances), ADP (124; 19% instances), ADV (91; 14% instances), CONJ (38; 6% instances), PART (27; 4% instances), VERB (13; 2% instances), ADJ (12; 2% instances), NOUN (12; 2% instances), PRON (10; 2% instances), PROPN (2; 0% instances), AUX (1; 0% instances), SCONJ (1; 0% instances), X (1; 0% instances)

---

Treebank Statistics (UD_Slovenian-SST)

There are 53 NUM lemmas (1%), 83 NUM types (1%) and 499 NUM tokens (2%). Out of 16 observed tags, the rank of NUM is: 8 in number of lemmas, 9 in number of types and 16 in number of tokens.

The 10 most frequent NUM lemmas: en, dva, trije, štirje, tisoč, dvajset, pet, petnajst, deset, sto

The 10 most frequent NUM types: eno, dva, en, ena, tri, tisoč, dvajset, dve, pet, enega

The 10 most frequent ambiguous lemmas: pet (NUM 17, ADJ 2)

The 10 most frequent ambiguous types: pet (NUM 14, X 1), ene (NUM 10, ADV 6), osemdesetih (ADJ 2, NUM 1)

• pet
  • NUM 14: pol sem jaz pol kar naenkrat imel ne vem pet jurjev naročnine
  • X 1: aha pet [gap] [gap] [all:laughter] … resno ?
• ene
  • NUM 10: in to so mešali kar eno olje pa ene take č [gap] stare barve
  • ADV 6: in mi je vsako jutro že za zajtrk skuhala po ene pet jajc pa masonek
• osemdesetih
  • ADJ 2: tole so mister mister broken wings bili so to časi osemdesetih let ko se je pač tale pesem držala pa še vedno se
  • NUM 1: v nadaljevanju pesem iz osemdesetih belinda karlyle heaven is a place on earth pa star čebelji pregovor tudi sledi

Morphology

The form / lemma ratio of NUM is 1.566038 (the average of all parts of speech is 1.575031).

The 1st highest number of forms (9) was observed with the lemma “en”: en, ena, ene, enega, enemu, eni, enih, enim, eno.

The 2nd highest number of forms (4) was observed with the lemma “trije”: treh, tremi, tri, trije.

The 3rd highest number of forms (4) was observed with the lemma “štirje”: štiri, štirih, štirim, štirje.

NUM occurs with 5 features: sl-feat/Case (499; 100% instances), sl-feat/NumForm (499; 100% instances), sl-feat/NumType (499; 100% instances), sl-feat/Number (499; 100% instances), sl-feat/Gender (270; 54% instances)

NUM occurs with 15 feature-value pairs: Case=Acc, Case=Dat, Case=Gen, Case=Ins, Case=Loc, Case=Nom, Gender=Fem, Gender=Masc, Gender=Neut, NumForm=Word, NumType=Card, NumType=Sets, Number=Dual, Number=Plur, Number=Sing

NUM occurs with 39 feature combinations. The most frequent feature combination is Case=Acc|Number=Plur|NumForm=Word|NumType=Card (122 tokens). Examples: dvajset, sto, tisoč, petnajst, deset, pet, šestdeset, osemdeset, petdeset, petsto

Relations

NUM nodes are attached to their parents using 15 different relations: sl-dep/nummod (284; 57% instances), sl-dep/compound (53; 11% instances), sl-dep/conj (30; 6% instances), sl-dep/nmod (29; 6% instances), sl-dep/root (26; 5% instances), sl-dep/nsubj (25; 5% instances), sl-dep/dobj (20; 4% instances), sl-dep/parataxis (9; 2% instances), sl-dep/reparandum (8; 2% instances), sl-dep/advmod (4; 1% instances), sl-dep/mwe (4; 1% instances), sl-dep/appos (2; 0% instances), sl-dep/ccomp (2; 0% instances), sl-dep/dislocated (2; 0% instances), sl-dep/acl (1; 0% instances)

Parents of NUM nodes belong to 11 different parts of speech: NOUN (271; 54% instances), NUM (87; 17% instances), VERB (81; 16% instances), ROOT (26; 5% instances), ADJ (12; 2% instances), PRON (12; 2% instances), PROPN (4; 1% instances), ADV (2; 0% instances), AUX (2; 0% instances), INTJ (1; 0% instances), X (1; 0% instances)

358 (72%) NUM nodes are leaves.

71 (14%) NUM nodes have one child.

33 (7%) NUM nodes have two children.

37 (7%) NUM nodes have three or more children.

The highest child degree of a NUM node is 8.

Children of NUM nodes are attached using 23 different relations: sl-dep/compound (54; 19% instances), sl-dep/advmod (42; 15% instances), sl-dep/conj (40; 14% instances), sl-dep/case (32; 11% instances), sl-dep/cc (17; 6% instances), sl-dep/cop (11; 4% instances), sl-dep/discourse (11; 4% instances), sl-dep/reparandum (10; 4% instances), sl-dep/nsubj (9; 3% instances), sl-dep/discourse:filler (7; 3% instances), sl-dep/nmod (7; 3% instances), sl-dep/parataxis (7; 3% instances), sl-dep/punct (7; 3% instances), sl-dep/acl (4; 1% instances), sl-dep/amod (4; 1% instances), sl-dep/det (4; 1% instances), sl-dep/mwe (4; 1% instances), sl-dep/mark (3; 1% instances), sl-dep/parataxis:discourse (2; 1% instances), sl-dep/aux (1; 0% instances), sl-dep/cc:preconj (1; 0% instances), sl-dep/expl (1; 0% instances), sl-dep/nummod (1; 0% instances)

Children of NUM nodes belong to 16 different parts of speech: NUM (87; 31% instances), ADP (37; 13% instances), ADV (28; 10% instances), VERB (22; 8% instances), PART (20; 7% instances), NOUN (16; 6% instances), PRON (15; 5% instances), CONJ (12; 4% instances), INTJ (10; 4% instances), X (10; 4% instances), ADJ (9; 3% instances), DET (4; 1% instances), PROPN (3; 1% instances), SCONJ (3; 1% instances), PUNCT (2; 1% instances), AUX (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]

© 2014 Universal Dependencies contributors. Site powered by Annodoc and brat

.