home edit page issue tracker

This page pertains to UD version 2.

Treebank Statistics: UD_English-Atis: POS Tags: ADJ

There are 76 ADJ lemmas (8%), 85 ADJ types (8%) and 2079 ADJ tokens (3%). Out of 14 observed tags, the rank of ADJ is: 5 in number of lemmas, 5 in number of types and 8 in number of tokens.

The 10 most frequent ADJ lemmas: one, cheap, early, available, next, seven, late, like, many, expensive

The 10 most frequent ADJ types: first, cheapest, available, earliest, next, seventh, early, latest, like, many

The 10 most frequent ambiguous lemmas: one (ADJ 240, NUM 178, PRON 4), cheap (ADJ 230, NOUN 1), early (ADJ 221, ADV 21), seven (ADJ 93, NUM 1), late (ADJ 91, ADV 3), like (VERB 380, ADP 88, ADJ 66, DET 1), nonstop (NOUN 129, ADJ 30, ADV 1), two (ADJ 29, NUM 19), much (ADJ 28, ADV 25), eight (ADJ 27, NUM 2)

The 10 most frequent ambiguous types: first (ADJ 240, ADV 27), earliest (ADJ 141, ADV 6), early (ADJ 78, ADV 15), like (VERB 380, ADP 88, ADJ 66, DET 1), nonstop (NOUN 127, ADJ 30, ADV 1), much (ADJ 28, ADV 25), less (ADJ 27, ADV 3), third (ADJ 26, PROPN 3), other (ADJ 15, NOUN 1), thirtieth (ADJ 10, NOUN 3)

Morphology

The form / lemma ratio of ADJ is 1.118421 (the average of all parts of speech is 1.144766).

The 1st highest number of forms (3) was observed with the lemma “early”: earlier, earliest, early.

The 2nd highest number of forms (3) was observed with the lemma “late”: late, later, latest.

The 3rd highest number of forms (2) was observed with the lemma “cheap”: cheap, cheapest.

ADJ occurs with 2 features: Degree (2079; 100% instances), NumType (521; 25% instances)

ADJ occurs with 4 feature-value pairs: Degree=Cmp, Degree=Pos, Degree=Sup, NumType=Ord

ADJ occurs with 4 feature combinations. The most frequent feature combination is Degree=Pos (968 tokens). Examples: available, first, next, early, like, many, expensive, daily, seventh, last

Relations

ADJ nodes are attached to their parents using 21 different relations: amod (1690; 81% instances), root (202; 10% instances), advmod (70; 3% instances), compound (28; 1% instances), nmod:tmod (13; 1% instances), ccomp (11; 1% instances), parataxis (11; 1% instances), acl:relcl (10; 0% instances), conj (9; 0% instances), nmod (6; 0% instances), nsubj (6; 0% instances), obl (4; 0% instances), advcl (3; 0% instances), appos (3; 0% instances), fixed (3; 0% instances), acl (2; 0% instances), discourse (2; 0% instances), obl:tmod (2; 0% instances), xcomp (2; 0% instances), dislocated (1; 0% instances), flat (1; 0% instances)

Parents of ADJ nodes belong to 11 different parts of speech: NOUN (1719; 83% instances), (202; 10% instances), VERB (57; 3% instances), ADJ (56; 3% instances), PROPN (23; 1% instances), ADV (7; 0% instances), NUM (5; 0% instances), PRON (4; 0% instances), ADP (2; 0% instances), AUX (2; 0% instances), DET (2; 0% instances)

1468 (71%) ADJ nodes are leaves.

320 (15%) ADJ nodes have one child.

51 (2%) ADJ nodes have two children.

240 (12%) ADJ nodes have three or more children.

The highest child degree of a ADJ node is 6.

Children of ADJ nodes are attached using 25 different relations: obl (298; 23% instances), nsubj (228; 17% instances), cop (158; 12% instances), advmod (150; 11% instances), compound (126; 10% instances), aux (75; 6% instances), det (46; 4% instances), obl:tmod (43; 3% instances), obj (34; 3% instances), xcomp (34; 3% instances), nmod (27; 2% instances), case (22; 2% instances), cc (10; 1% instances), conj (10; 1% instances), mark (10; 1% instances), nmod:tmod (9; 1% instances), parataxis (9; 1% instances), amod (8; 1% instances), discourse (5; 0% instances), flat (3; 0% instances), nummod (3; 0% instances), advcl (2; 0% instances), dep (1; 0% instances), dislocated (1; 0% instances), expl (1; 0% instances)

Children of ADJ nodes belong to 12 different parts of speech: NOUN (319; 24% instances), AUX (235; 18% instances), PROPN (235; 18% instances), NUM (129; 10% instances), ADV (115; 9% instances), PRON (84; 6% instances), ADJ (56; 4% instances), VERB (49; 4% instances), DET (47; 4% instances), ADP (29; 2% instances), CCONJ (10; 1% instances), INTJ (5; 0% instances)