Treebank Statistics: UD_Welsh-CCG: POS Tags: NUM
There are 228 NUM
lemmas (4%), 249 NUM
types (3%) and 689 NUM
tokens (1%).
Out of 15 observed tags, the rank of NUM
is: 5 in number of lemmas, 5 in number of types and 13 in number of tokens.
The 10 most frequent NUM
lemmas: un, dau, tri, pedwar, chwech, 4, pump, 2019, ugain, deg
The 10 most frequent NUM
types: un, chwe, 4, dau, ddau, ddwy, tri, tair, pedair, 2019
The 10 most frequent ambiguous lemmas: 2 (NUM 3, PROPN 1), 70 (NUM 3, NOUN 2), miliwn (NOUN 7, NUM 3), 1950 (NUM 2, NOUN 1), 90 (NUM 2, NOUN 1), cant (NOUN 10, NUM 2), neb (PRON 8, NUM 2), 1930 (NOUN 1, NUM 1), 1970 (NOUN 4, NUM 1), 1990 (NOUN 2, NUM 1)
The 10 most frequent ambiguous types: dwy (NUM 6, VERB 1), 17 (NUM 3, ADJ 2), 2 (NUM 3, PROPN 1), I (ADP 15, NUM 3, PRON 1), deg (NUM 3, ADJ 2), 13 (NUM 2, ADJ 1), 18 (ADJ 2, NUM 2), 19 (ADJ 3, NUM 2), chan (NUM 2, ADP 1), neb (PRON 8, NUM 2)
- dwy
- 17
- 2
- I
- deg
- 13
- 18
- 19
- chan
- NUM 2: Gall organ hynafol Eglwys Dewi Sant , Glanadda fynd i goleg dri chan milltir o Fangor .
- ADP 1: Aelod o ‘r gangen Frythonaidd o ‘r ieithoedd Celtaidd a siaredir yn frodorol yng Nghymru , gan Gymry a phobl eraill ar wasgar yn Lloegr , a chan gymuned fechan yn Y Wladfa , yr Ariannin yw ‘r Gymraeg .
- neb
Morphology
The form / lemma ratio of NUM
is 1.092105 (the average of all parts of speech is 1.452021).
The 1st highest number of forms (6) was observed with the lemma “tri”: dair, dri, tair, thair, thri, tri.
The 2nd highest number of forms (5) was observed with the lemma “dau”: dau, ddau, ddwy, dwy, dyw.
The 3rd highest number of forms (5) was observed with the lemma “pedwar”: bedair, bedwar, pedair, pedwar, phedair.
NUM
occurs with 4 features: NumType (685; 99% instances), NumForm (682; 99% instances), Gender (91; 13% instances), Mutation (69; 10% instances)
NUM
occurs with 9 feature-value pairs: Gender=Fem
, Gender=Masc
, Mutation=AM
, Mutation=SM
, NumForm=Combi
, NumForm=Digit
, NumForm=Roman
, NumForm=Word
, NumType=Card
NUM
occurs with 15 feature combinations.
The most frequent feature combination is NumForm=Digit|NumType=Card
(333 tokens).
Examples: 4, 2019, 10, 200, 2020, 50, 500, 7, 100, 11
Relations
NUM
nodes are attached to their parents using 16 different relations: nummod (329; 48% instances), obl (123; 18% instances), nmod (108; 16% instances), nsubj (36; 5% instances), conj (28; 4% instances), obj (20; 3% instances), root (16; 2% instances), appos (9; 1% instances), acl (6; 1% instances), flat:name (4; 1% instances), amod (2; 0% instances), ccomp (2; 0% instances), flat (2; 0% instances), parataxis (2; 0% instances), acl:relcl (1; 0% instances), advcl (1; 0% instances)
Parents of NUM
nodes belong to 9 different parts of speech: NOUN (487; 71% instances), VERB (86; 12% instances), SYM (40; 6% instances), NUM (24; 3% instances), ADJ (16; 2% instances), (16; 2% instances), PROPN (15; 2% instances), PRON (4; 1% instances), AUX (1; 0% instances)
393 (57%) NUM
nodes are leaves.
153 (22%) NUM
nodes have one child.
84 (12%) NUM
nodes have two children.
59 (9%) NUM
nodes have three or more children.
The highest child degree of a NUM
node is 7.
Children of NUM
nodes are attached using 18 different relations: case (183; 33% instances), nmod (129; 23% instances), punct (47; 8% instances), cc (26; 5% instances), amod (24; 4% instances), det (24; 4% instances), conj (23; 4% instances), advmod (22; 4% instances), cop (21; 4% instances), case:pred (16; 3% instances), nsubj (13; 2% instances), acl:relcl (8; 1% instances), obl (5; 1% instances), acl (4; 1% instances), appos (4; 1% instances), nummod (4; 1% instances), xcomp (2; 0% instances), mark (1; 0% instances)
Children of NUM
nodes belong to 14 different parts of speech: ADP (185; 33% instances), NOUN (141; 25% instances), PUNCT (47; 8% instances), ADJ (28; 5% instances), CCONJ (26; 5% instances), NUM (24; 4% instances), AUX (21; 4% instances), DET (21; 4% instances), PART (21; 4% instances), PRON (16; 3% instances), ADV (12; 2% instances), VERB (7; 1% instances), PROPN (5; 1% instances), SYM (2; 0% instances)