home edit page issue tracker

This page pertains to UD version 2.

Treebank Statistics: UD_Russian-Taiga: POS Tags: SCONJ

There are 43 SCONJ lemmas (0%), 47 SCONJ types (0%) and 3506 SCONJ tokens (2%). Out of 17 observed tags, the rank of SCONJ is: 15 in number of lemmas, 15 in number of types and 12 in number of tokens.

The 10 most frequent SCONJ lemmas: что, как, если, когда, чтобы, то, хотя, чем, чтоб, пока

The 10 most frequent SCONJ types: что, как, если, когда, чтобы, то, хотя, чем, чтоб, пока

The 10 most frequent ambiguous lemmas: что (SCONJ 1152, PRON 821, ADV 12, PART 7), как (SCONJ 732, ADV 476, CCONJ 1, PART 1), когда (SCONJ 270, ADV 19), то (PRON 464, SCONJ 185, PART 66, CCONJ 54, DET 1), хотя (SCONJ 96, PART 39), чем (SCONJ 90, PRON 3), пока (ADV 62, SCONJ 45), хоть (PART 33, SCONJ 28), ведь (PART 55, SCONJ 20), будто (PART 39, SCONJ 13)

The 10 most frequent ambiguous types: что (SCONJ 1150, PRON 498, ADV 13, PART 6, X 1), как (SCONJ 657, ADV 320, X 3, CCONJ 1, DET 1, PART 1), когда (SCONJ 216, ADV 20, X 1), чтобы (SCONJ 181, PRON 1), то (SCONJ 176, PRON 174, X 127, PART 65, CCONJ 49, DET 23, ADV 1), хотя (SCONJ 68, PART 36), чем (PRON 90, SCONJ 87, X 1), пока (ADV 57, SCONJ 32), хоть (PART 30, SCONJ 17), ведь (PART 37, SCONJ 17)

Morphology

The form / lemma ratio of SCONJ is 1.093023 (the average of all parts of speech is 1.879397).

The 1st highest number of forms (3) was observed with the lemma “чтоб”: Чтоп, что, чтоб.

The 2nd highest number of forms (3) was observed with the lemma “чтобы”: что, чтоб, чтобы.

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

SCONJ occurs with 3 features: Mood (282; 8% instances), Typo (25; 1% instances), Abbr (14; 0% instances)

SCONJ occurs with 3 feature-value pairs: Abbr=Yes, Mood=Cnd, Typo=Yes

SCONJ occurs with 5 feature combinations. The most frequent feature combination is _ (3208 tokens). Examples: что, как, если, когда, то, хотя, чем, пока, хоть, ведь

Relations

SCONJ nodes are attached to their parents using 14 different relations: mark (2963; 85% instances), case (277; 8% instances), fixed (184; 5% instances), cc (32; 1% instances), advmod (25; 1% instances), parataxis (8; 0% instances), nsubj (4; 0% instances), obl (4; 0% instances), obj (3; 0% instances), root (2; 0% instances), acl (1; 0% instances), acl:relcl (1; 0% instances), conj (1; 0% instances), reparandum (1; 0% instances)

Parents of SCONJ nodes belong to 16 different parts of speech: VERB (2111; 60% instances), NOUN (600; 17% instances), ADJ (319; 9% instances), ADV (261; 7% instances), PRON (72; 2% instances), PROPN (53; 2% instances), NUM (23; 1% instances), DET (20; 1% instances), AUX (15; 0% instances), PART (14; 0% instances), X (5; 0% instances), INTJ (4; 0% instances), SCONJ (3; 0% instances), ADP (2; 0% instances), (2; 0% instances), SYM (2; 0% instances)

3395 (97%) SCONJ nodes are leaves.

101 (3%) SCONJ nodes have one child.

6 (0%) SCONJ nodes have two children.

4 (0%) SCONJ nodes have three or more children.

The highest child degree of a SCONJ node is 5.

Children of SCONJ nodes are attached using 9 different relations: fixed (78; 60% instances), goeswith (22; 17% instances), punct (16; 12% instances), advmod (7; 5% instances), nsubj (2; 2% instances), obl (2; 2% instances), conj (1; 1% instances), obj (1; 1% instances), reparandum (1; 1% instances)

Children of SCONJ nodes belong to 13 different parts of speech: PART (47; 36% instances), NOUN (24; 18% instances), X (22; 17% instances), PUNCT (16; 12% instances), PRON (8; 6% instances), ADV (3; 2% instances), SCONJ (3; 2% instances), VERB (2; 2% instances), ADJ (1; 1% instances), ADP (1; 1% instances), AUX (1; 1% instances), CCONJ (1; 1% instances), PROPN (1; 1% instances)