This is part of archived UD v1 documentation. See http://universaldependencies.org/ for the current version.
home cs/dep issue tracker

punct: punctuation

This is used for any piece of punctuation in a clause. See

Tokens with the relation punct always attach to content words (except in cases of ellipsis) and can never have dependents. Since punct is not a normal dependency relation, the usual criteria for determining the head word do not apply. Instead, we use the following principles:

  1. A punctuation mark separating coordinated units is attached to the first conjunct.
  2. A punctuation mark preceding or following a subordinated unit is attached to this unit.
  3. Within the relevant unit, a punctuation mark is attached at the highest possible node that preserves projectivity.
  4. Paired punctuation marks (quotes and brackets) should be attached to the same word unless that would create non-projectivity. This word is usually the head of the phrase enclosed in the paired punctuation.

Treebank Statistics (UD_Czech)

This relation is universal.

220974 nodes (15%) are attached to their parents as punct.

156731 instances of punct (71%) are left-to-right (parent precedes child). Average distance between parent and child is 7.24895236543666.

The following 18 pairs of parts of speech are connected with punct: VERB-PUNCT (126081; 57% instances), NOUN-PUNCT (40358; 18% instances), PROPN-PUNCT (19518; 9% instances), ADJ-PUNCT (17741; 8% instances), NUM-PUNCT (11542; 5% instances), ADV-PUNCT (2977; 1% instances), PRON-PUNCT (1657; 1% instances), PART-PUNCT (515; 0% instances), PUNCT-PUNCT (158; 0% instances), DET-PUNCT (129; 0% instances), CONJ-PUNCT (73; 0% instances), SYM-PUNCT (72; 0% instances), INTJ-PUNCT (68; 0% instances), ADP-PUNCT (40; 0% instances), SCONJ-PUNCT (31; 0% instances), X-PUNCT (11; 0% instances), AUX-PUNCT (2; 0% instances), NOUN-PROPN (1; 0% instances).


Treebank Statistics (UD_Czech-CAC)

This relation is universal.

60122 nodes (12%) are attached to their parents as punct.

43952 instances of punct (73%) are left-to-right (parent precedes child). Average distance between parent and child is 9.08610824656532.

The following 18 pairs of parts of speech are connected with punct: VERB-PUNCT (37539; 62% instances), NOUN-PUNCT (13327; 22% instances), ADJ-PUNCT (5561; 9% instances), PROPN-PUNCT (1439; 2% instances), ADV-PUNCT (925; 2% instances), PRON-PUNCT (625; 1% instances), NUM-PUNCT (323; 1% instances), SYM-PUNCT (200; 0% instances), PART-PUNCT (68; 0% instances), SCONJ-PUNCT (50; 0% instances), DET-PUNCT (28; 0% instances), CONJ-PUNCT (11; 0% instances), AUX-PUNCT (10; 0% instances), INTJ-PUNCT (9; 0% instances), ADJ-SYM (2; 0% instances), ADP-PUNCT (2; 0% instances), PUNCT-PUNCT (2; 0% instances), SYM-SYM (1; 0% instances).


Treebank Statistics (UD_Czech-CLTT)

This relation is universal.

3991 nodes (11%) are attached to their parents as punct.

2347 instances of punct (59%) are left-to-right (parent precedes child). Average distance between parent and child is 17.1911801553495.

The following 11 pairs of parts of speech are connected with punct: VERB-PUNCT (2157; 54% instances), NOUN-PUNCT (1273; 32% instances), ADJ-PUNCT (314; 8% instances), X-PUNCT (125; 3% instances), NUM-PUNCT (86; 2% instances), ADV-PUNCT (16; 0% instances), PART-PUNCT (8; 0% instances), PRON-PUNCT (8; 0% instances), SYM-PUNCT (2; 0% instances), DET-PUNCT (1; 0% instances), SCONJ-PUNCT (1; 0% instances).


punct in other languages: [bg] [cs] [de] [el] [en] [es] [eu] [fa] [fi] [fr] [ga] [he] [hu] [it] [ja] [ko] [sv] [u]
BESbswyBESbswyBESbswyBESbswy