home edit page issue tracker

This page pertains to UD version 2.

Treebank Statistics: UD_Finnish-TDT: POS Tags: NUM

There are 1205 NUM lemmas (5%), 1376 NUM types (3%) and 4979 NUM tokens (2%). Out of 15 observed tags, the rank of NUM is: 6 in number of lemmas, 6 in number of types and 10 in number of tokens.

The 10 most frequent NUM lemmas: kaksi, yksi, 1, 2, kolme, 2009, 3, miljoona, viisi, neljä

The 10 most frequent NUM types: 1, 2, kaksi, 2009, 3, 5, 4, kolme, yksi, 10

The 10 most frequent ambiguous lemmas: yksi (NUM 198, PRON 52), 1 (NUM 163, PROPN 2), 2 (NUM 150, PROPN 1), kuusi (NUM 42, NOUN 7), 20 (NUM 40, PROPN 1), pari (NUM 36, NOUN 9), puoli (NOUN 81, NUM 25), 25 (NUM 16, ADJ 1), miljardi (NUM 15, NOUN 3), 600 (NUM 4, PROPN 2)

The 10 most frequent ambiguous types: 1 (NUM 160, PROPN 1), yksi (NUM 56, PRON 28), yhden (NUM 51, PRON 3), 20 (NUM 39, PROPN 1), pari (NUM 19, NOUN 2), kuusi (NUM 14, NOUN 2), yhtä (ADV 49, NUM 13, PRON 4), puoli (NUM 13, NOUN 6), parin (NUM 11, NOUN 2), yhdessä (ADV 35, NUM 7)

Morphology

The form / lemma ratio of NUM is 1.141909 (the average of all parts of speech is 2.067974).

The 1st highest number of forms (18) was observed with the lemma “yksi”: yhdeksi, yhdelle, yhdellä, yhdeltä, yhdeltäkään, yhden, yhdessä, yhdessäkään, yhdestä, yhdet, yhteen, yhtenä, yhtenäkin, yhtä, yhtäkään, yks, yksi, yksikin.

The 2nd highest number of forms (15) was observed with the lemma “kaksi”: kahdella, kahdelle, kahdelta, kahden, kahdessa, kahdesta, kahdet, kahta, kahteen, kahtena, kaks, kaksi, kaksi-, kaksiin, kaksin.

The 3rd highest number of forms (12) was observed with the lemma “tuhat”: tuhannella, tuhannen, tuhannesta, tuhannet, tuhansia, tuhansien, tuhansiin, tuhansilta, tuhansista, tuhanteenkin, tuhat, tuhatta.

NUM occurs with 8 features: NumType (4684; 94% instances), Case (1285; 26% instances), Number (1285; 26% instances), Style (13; 0% instances), Clitic (11; 0% instances), Typo (4; 0% instances), Abbr (2; 0% instances), PronType (1; 0% instances)

NUM occurs with 21 feature-value pairs: Abbr=Yes, Case=Abl, Case=Ade, Case=All, Case=Ela, Case=Ess, Case=Gen, Case=Ill, Case=Ine, Case=Ins, Case=Nom, Case=Par, Case=Tra, Clitic=Kaan, Clitic=Kin, NumType=Card, Number=Plur, Number=Sing, PronType=Ind, Style=Coll, Typo=Yes

NUM occurs with 45 feature combinations. The most frequent feature combination is NumType=Card (3431 tokens). Examples: 1, 2, 2009, 3, 5, 4, kaksi, 10, 2006, 6

Relations

NUM nodes are attached to their parents using 21 different relations: nummod (3286; 66% instances), obl (576; 12% instances), flat (358; 7% instances), appos (176; 4% instances), compound (158; 3% instances), conj (124; 2% instances), nmod (101; 2% instances), flat:name (50; 1% instances), root (49; 1% instances), nsubj (23; 0% instances), nmod:poss (18; 0% instances), obj (17; 0% instances), nsubj:cop (15; 0% instances), orphan (7; 0% instances), ccomp (6; 0% instances), advcl (5; 0% instances), acl:relcl (4; 0% instances), compound:nn (3; 0% instances), amod (1; 0% instances), nmod:gsubj (1; 0% instances), parataxis (1; 0% instances)

Parents of NUM nodes belong to 10 different parts of speech: NOUN (3423; 69% instances), VERB (569; 11% instances), NUM (424; 9% instances), ADJ (260; 5% instances), PROPN (157; 3% instances), SYM (74; 1% instances), (49; 1% instances), ADV (13; 0% instances), PRON (7; 0% instances), X (3; 0% instances)

3566 (72%) NUM nodes are leaves.

752 (15%) NUM nodes have one child.

429 (9%) NUM nodes have two children.

232 (5%) NUM nodes have three or more children.

The highest child degree of a NUM node is 9.

Children of NUM nodes are attached using 25 different relations: punct (889; 36% instances), advmod (440; 18% instances), flat (389; 16% instances), compound (147; 6% instances), conj (119; 5% instances), cc (92; 4% instances), nsubj:cop (69; 3% instances), cop (64; 3% instances), nmod (61; 2% instances), appos (49; 2% instances), obl (36; 1% instances), case (22; 1% instances), amod (16; 1% instances), det (15; 1% instances), orphan (13; 1% instances), mark (11; 0% instances), acl:relcl (8; 0% instances), aux (7; 0% instances), flat:name (6; 0% instances), advcl (5; 0% instances), nummod (4; 0% instances), acl (2; 0% instances), goeswith (2; 0% instances), discourse (1; 0% instances), nmod:poss (1; 0% instances)

Children of NUM nodes belong to 15 different parts of speech: PUNCT (889; 36% instances), ADV (436; 18% instances), NUM (424; 17% instances), NOUN (405; 16% instances), CCONJ (92; 4% instances), AUX (71; 3% instances), ADJ (41; 2% instances), PRON (33; 1% instances), PROPN (24; 1% instances), ADP (22; 1% instances), SCONJ (11; 0% instances), VERB (10; 0% instances), SYM (7; 0% instances), X (2; 0% instances), INTJ (1; 0% instances)