home edit page issue tracker

This page pertains to UD version 2.

Treebank Statistics: UD_Latvian-LVTB: POS Tags: NUM

There are 600 NUM lemmas (3%), 661 NUM types (1%) and 3562 NUM tokens (1%). Out of 17 observed tags, the rank of NUM is: 7 in number of lemmas, 7 in number of types and 14 in number of tokens.

The 10 most frequent NUM lemmas: viens, divi, trīs, otrs, četri, pieci, 20, desmit, 3, seši

The 10 most frequent NUM types: viens, trīs, vienu, viena, divas, vienā, divi, 20, desmit, otru

The 10 most frequent ambiguous lemmas: viens (NUM 622, ADJ 2, PRON 1), otrs (NUM 147, ADJ 7), pusotrs (NUM 6, ADJ 4, NOUN 3), i (PART 5, CCONJ 3, NUM 1, SYM 1), tūkstotis (NOUN 48, NUM 5), II (NUM 4, ADJ 3), III (ADJ 9, NUM 4), 1. (ADJ 62, NUM 2), desmits (NOUN 16, NUM 2), ex (NUM 2, X 2)

The 10 most frequent ambiguous types: vienu (NUM 147, PRON 1, X 1), otrā (ADJ 19, NUM 17), 8 (NUM 17, X 1), vieni (NUM 5, PRON 1), I (ADJ 5, NUM 5, CCONJ 2), II (NUM 4, ADJ 3), III (ADJ 9, NUM 4), pusotra (NUM 3, ADJ 1), pusotru (NOUN 4, NUM 3, ADJ 1), 1. (ADJ 62, NUM 2)

Morphology

The form / lemma ratio of NUM is 1.101667 (the average of all parts of speech is 2.244795).

The 1st highest number of forms (11) was observed with the lemma “viens”: Vienām, viena, vienai, vienam, vienas, vieni, vieniem, vienos, viens, vienu, vienā.

The 2nd highest number of forms (10) was observed with the lemma “otrs”: otra, otrai, otram, otras, otriem, otrs, otru, otrā, otrām, otrās.

The 3rd highest number of forms (8) was observed with the lemma “divi”: divas, divi, diviem, divos, divu, divus, divām, divās.

NUM occurs with 5 features: NumType (3562; 100% instances), Number (1799; 51% instances), Gender (1670; 47% instances), Case (1664; 47% instances), Typo (4; 0% instances)

NUM occurs with 12 feature-value pairs: Case=Acc, Case=Dat, Case=Gen, Case=Loc, Case=Nom, Gender=Fem, Gender=Masc, NumType=Card, NumType=Frac, Number=Plur, Number=Sing, Typo=Yes

NUM occurs with 32 feature combinations. The most frequent feature combination is NumType=Card (1761 tokens). Examples: viens, trīs, vienu, viena, 20, 3, 2, 1, 15, 30

Relations

NUM nodes are attached to their parents using 24 different relations: nummod (2631; 74% instances), conj (187; 5% instances), parataxis (119; 3% instances), nsubj (90; 3% instances), dep (84; 2% instances), root (69; 2% instances), nmod (58; 2% instances), flat:name (50; 1% instances), compound (49; 1% instances), obj (42; 1% instances), iobj (41; 1% instances), obl (40; 1% instances), xcomp (35; 1% instances), flat (17; 0% instances), discourse (11; 0% instances), nsubj:pass (7; 0% instances), acl (6; 0% instances), appos (6; 0% instances), ccomp (5; 0% instances), orphan (5; 0% instances), advcl (4; 0% instances), flat:foreign (4; 0% instances), amod (1; 0% instances), csubj (1; 0% instances)

Parents of NUM nodes belong to 12 different parts of speech: NOUN (2433; 68% instances), VERB (437; 12% instances), NUM (241; 7% instances), SYM (205; 6% instances), PROPN (74; 2% instances), (69; 2% instances), X (47; 1% instances), ADJ (33; 1% instances), ADV (13; 0% instances), PRON (8; 0% instances), AUX (1; 0% instances), CCONJ (1; 0% instances)

2335 (66%) NUM nodes are leaves.

812 (23%) NUM nodes have one child.

241 (7%) NUM nodes have two children.

174 (5%) NUM nodes have three or more children.

The highest child degree of a NUM node is 11.

Children of NUM nodes are attached using 27 different relations: punct (573; 28% instances), nmod (248; 12% instances), advmod (247; 12% instances), conj (194; 9% instances), case (163; 8% instances), discourse (89; 4% instances), cc (79; 4% instances), cop (78; 4% instances), nsubj (66; 3% instances), obl (56; 3% instances), compound (53; 3% instances), flat:name (33; 2% instances), det (29; 1% instances), dep (28; 1% instances), acl (25; 1% instances), amod (22; 1% instances), flat (21; 1% instances), orphan (16; 1% instances), parataxis (14; 1% instances), mark (9; 0% instances), advcl (7; 0% instances), csubj (7; 0% instances), nummod (7; 0% instances), flat:foreign (6; 0% instances), appos (4; 0% instances), iobj (2; 0% instances), goeswith (1; 0% instances)

Children of NUM nodes belong to 16 different parts of speech: PUNCT (573; 28% instances), NOUN (317; 15% instances), ADV (293; 14% instances), NUM (241; 12% instances), ADP (159; 8% instances), PART (82; 4% instances), AUX (78; 4% instances), PRON (68; 3% instances), CCONJ (66; 3% instances), VERB (48; 2% instances), PROPN (35; 2% instances), ADJ (34; 2% instances), DET (28; 1% instances), SCONJ (26; 1% instances), SYM (22; 1% instances), X (7; 0% instances)