X: other
Definition
The tag X is used for words that for some reason cannot be assigned
a real part-of-speech category.
The X tag is disprefered, but kept in the treebank from the conversion of the previous tags in the initial distribution of the Copenhagen Dependency Treebank for foreign words, OCR errors, and parts of distributionally arranged compounds like musik- og billedprogrammer “music and image programs”
Treebank Statistics (UD_Danish)
There are 366 X lemmas (3%), 365 X types (2%) and 439 X tokens (0%).
Out of 17 observed tags, the rank of X is: 7 in number of lemmas, 7 in number of types and 15 in number of tokens.
The 10 most frequent X lemmas: of, MEDARB, en, vivaldi, we, allesammen, are, at, frue, hel
The 10 most frequent X types: of, MEDARB, en, vivaldi, we, allesammen, are, at, fru, hel
The 10 most frequent ambiguous lemmas: en (DET 2187, PRON 104, X 7), at (PART 1215, SCONJ 957, X 3), hel (ADJ 98, X 3), med (ADP 1045, ADV 82, X 3), the (X 3, NOUN 1), af (ADP 1265, ADV 29, X 2), fremfor (X 2, ADP 1), følge (VERB 40, NOUN 12, X 2), i (ADP 2838, ADV 10, X 1), in (X 2, ADJ 1)
The 10 most frequent ambiguous types: en (DET 1395, PRON 49, X 7), at (PART 1209, SCONJ 943, X 3), hel (ADJ 12, X 3), med (ADP 1008, ADV 80, X 3), the (X 3, NOUN 1), af (ADP 1251, ADV 29, X 2), forsknings- (NOUN 2, X 1), fremfor (X 2, ADP 1), følge (NOUN 11, VERB 8, X 2), i (ADP 2622, ADV 10, X 1)
- en
- at
- hel
- med
- the
- af
- forsknings-
- NOUN 2: Bemærkninger til forslag til folketingsbeslutning om et bioteknisk forsknings- og udviklingsprogram , fremsat 21. marts 1986 af Jytte Hilden m.fl. .
- X 1: Men nu får vi altså en chance , både ESA og NASA åbner mulighed for at få små robuste forsknings- og undervisningsdrabanter med som blaffere ved de kommercielle opsendelser , “ siger John Jørgensen .
- fremfor
- X 2: ” Da vi begyndte indspilningen af “ Chicism “ , havde vi egentlig tænkt os , at vi ville bruge hip-hop-genren som udgangspunkt , men vi fandt hurtigt ud af , at vi er bedre til at spille musik med rigtig instrumentering , fremfor at bruge tonsvis af trommemaskiner og computer-sampling .
- ADP 1: Det er linjen fra vælgerne til de styrende , der er slidt så tynd , at de fleste nu selv må hente viden i medierne og miljøet fremfor at følge politikerne - også i meget svære sager .
- følge
- NOUN 11: Hertil kommer besparelser som følge af reduceret fravær på arbejdet m.v. .
- VERB 8: Biologer udstyrer truede dyr med radiosendere , så de kan følge dem konstant og få viden , der kan redde dyrene
- X 2: ” Den udtalelse er i følge Jan Carlzon selv fremsat under en diskussion om faren for , at medier drager for hastige konklusioner .
- i
Morphology
The form / lemma ratio of X is 0.997268 (the average of all parts of speech is 1.355873).
The 1st highest number of forms (1) was observed with the lemma “04099l”: 04099l.
The 2nd highest number of forms (1) was observed with the lemma “12all”: 12all.
The 3rd highest number of forms (1) was observed with the lemma “16-ventilet”: 16-ventilet.
X occurs with 2 features: Foreign (111; 25% instances), Abbr (32; 7% instances)
X occurs with 2 feature-value pairs: Abbr=Yes, Foreign=Yes
X occurs with 3 feature combinations.
The most frequent feature combination is _ (296 tokens).
Examples: MEDARB, en, vivaldi, allesammen, at, hel, med, Statskundskab/Samfundsfag, af, aller
Relations
X nodes are attached to their parents using 17 different relations: advmod (95; 22% instances), nmod (67; 15% instances), foreign (61; 14% instances), amod (45; 10% instances), conj (41; 9% instances), dobj (25; 6% instances), name (21; 5% instances), nsubj (21; 5% instances), root (21; 5% instances), list (13; 3% instances), nmod:poss (10; 2% instances), appos (6; 1% instances), cc (4; 1% instances), mark (4; 1% instances), nmod:tmod (2; 0% instances), nsubjpass (2; 0% instances), nmod:loc (1; 0% instances)
Parents of X nodes belong to 13 different parts of speech: NOUN (113; 26% instances), VERB (94; 21% instances), X (75; 17% instances), PROPN (47; 11% instances), NUM (33; 8% instances), ADJ (21; 5% instances), ROOT (21; 5% instances), PRON (17; 4% instances), ADV (8; 2% instances), AUX (4; 1% instances), PART (3; 1% instances), ADP (2; 0% instances), CONJ (1; 0% instances)
229 (52%) X nodes are leaves.
89 (20%) X nodes have one child.
63 (14%) X nodes have two children.
58 (13%) X nodes have three or more children.
The highest child degree of a X node is 10.
Children of X nodes are attached using 21 different relations: conj (76; 17% instances), punct (65; 15% instances), cc (58; 13% instances), foreign (56; 13% instances), case (51; 12% instances), nmod (38; 9% instances), dobj (15; 3% instances), amod (13; 3% instances), nsubj (12; 3% instances), name (9; 2% instances), advcl (8; 2% instances), advmod (8; 2% instances), nmod:poss (8; 2% instances), appos (5; 1% instances), list (5; 1% instances), mark (3; 1% instances), acl:relcl (2; 0% instances), nummod (2; 0% instances), expl (1; 0% instances), nmod:loc (1; 0% instances), vocative (1; 0% instances)
Children of X nodes belong to 13 different parts of speech: NOUN (104; 24% instances), X (75; 17% instances), PUNCT (65; 15% instances), ADP (51; 12% instances), CONJ (48; 11% instances), PROPN (34; 8% instances), ADJ (17; 4% instances), ADV (13; 3% instances), VERB (12; 3% instances), SYM (8; 2% instances), NUM (4; 1% instances), PRON (4; 1% instances), SCONJ (2; 0% instances)
X in other languages: [bg] [cs] [de] [el] [en] [es] [eu] [fa] [fi] [fr] [ga] [he] [hu] [it] [ja] [ko] [sv] [u]