home edit page issue tracker

This page pertains to UD version 2.

Treebank Statistics: UD_English-GUM: POS Tags: SCONJ

There are 59 SCONJ lemmas (0%), 59 SCONJ types (0%) and 2767 SCONJ tokens (2%). Out of 17 observed tags, the rank of SCONJ is: 10 in number of lemmas, 11 in number of types and 14 in number of tokens.

The 10 most frequent SCONJ lemmas: that, if, as, because, how, for, by, while, of, in

The 10 most frequent SCONJ types: that, if, as, because, how, for, by, while, of, in

The 10 most frequent ambiguous lemmas: that (PRON 868, SCONJ 607, DET 217, ADP 30, ADV 11), if (SCONJ 381, ADP 16), as (ADP 479, SCONJ 275, ADV 138), because (SCONJ 171, ADP 13), how (ADV 106, SCONJ 105, NOUN 6), for (ADP 1030, SCONJ 96), by (ADP 538, SCONJ 92, ADV 1), while (SCONJ 92, NOUN 14), of (ADP 4105, SCONJ 84, ADV 23), in (ADP 2859, SCONJ 83, ADV 17, X 5, NOUN 1)

The 10 most frequent ambiguous types: that (PRON 735, SCONJ 605, DET 145, ADP 30, ADV 11), if (SCONJ 239, ADP 16), as (ADP 446, SCONJ 239, ADV 131), because (SCONJ 152, ADP 12), how (SCONJ 96, ADV 51, NOUN 6), for (ADP 939, SCONJ 92, ADV 1), by (ADP 498, SCONJ 81, ADV 1), while (SCONJ 63, NOUN 14), of (ADP 4098, SCONJ 84, ADV 15), in (ADP 2570, SCONJ 73, ADV 17, X 5)

Morphology

The form / lemma ratio of SCONJ is 1.000000 (the average of all parts of speech is 1.226279).

The 1st highest number of forms (1) was observed with the lemma “a”: a.

The 2nd highest number of forms (1) was observed with the lemma “about”: about.

The 3rd highest number of forms (1) was observed with the lemma “after”: after.

SCONJ occurs with 4 features: PronType (172; 6% instances), Typo (5; 0% instances), Degree (3; 0% instances), VerbForm (2; 0% instances)

SCONJ occurs with 5 feature-value pairs: Degree=Pos, PronType=Int, PronType=Rel, Typo=Yes, VerbForm=Ger

SCONJ occurs with 6 feature combinations. The most frequent feature combination is _ (2585 tokens). Examples: that, if, as, because, for, by, while, of, in, after

Relations

SCONJ nodes are attached to their parents using 10 different relations: mark (2622; 95% instances), obj (73; 3% instances), obl (23; 1% instances), conj (20; 1% instances), nmod (18; 1% instances), ccomp (5; 0% instances), acl (2; 0% instances), compound (2; 0% instances), acl:relcl (1; 0% instances), obl:agent (1; 0% instances)

Parents of SCONJ nodes belong to 13 different parts of speech: VERB (2187; 79% instances), ADJ (268; 10% instances), NOUN (206; 7% instances), ADV (26; 1% instances), PRON (23; 1% instances), PROPN (14; 1% instances), NUM (13; 0% instances), AUX (11; 0% instances), SCONJ (10; 0% instances), DET (5; 0% instances), X (2; 0% instances), INTJ (1; 0% instances), PART (1; 0% instances)

2497 (90%) SCONJ nodes are leaves.

202 (7%) SCONJ nodes have one child.

49 (2%) SCONJ nodes have two children.

19 (1%) SCONJ nodes have three or more children.

The highest child degree of a SCONJ node is 5.

Children of SCONJ nodes are attached using 19 different relations: advcl:relcl (103; 28% instances), fixed (83; 23% instances), punct (68; 19% instances), case (41; 11% instances), acl (21; 6% instances), cc (12; 3% instances), conj (9; 2% instances), nsubj (6; 2% instances), advmod (5; 1% instances), cop (5; 1% instances), mark (3; 1% instances), aux (2; 1% instances), reparandum (2; 1% instances), discourse (1; 0% instances), nmod:poss (1; 0% instances), obl (1; 0% instances), obl:npmod (1; 0% instances), orphan (1; 0% instances), parataxis (1; 0% instances)

Children of SCONJ nodes belong to 13 different parts of speech: VERB (118; 32% instances), ADP (98; 27% instances), PUNCT (68; 19% instances), NOUN (30; 8% instances), CCONJ (12; 3% instances), SCONJ (10; 3% instances), ADV (9; 2% instances), AUX (9; 2% instances), PRON (5; 1% instances), ADJ (2; 1% instances), INTJ (2; 1% instances), PART (2; 1% instances), PROPN (1; 0% instances)