home edit page issue tracker

Treebank Statistics: UD_Breton-KEB: POS Tags: ADJ

There are 200 ADJ lemmas (11%), 228 ADJ types (9%) and 435 ADJ tokens (4%). Out of 17 observed tags, the rank of ADJ is: 3 in number of lemmas, 3 in number of types and 8 in number of tokens.

The 10 most frequent ADJ lemmas: bras, mat, ret, nevez, kentañ, yaouank, all, brav, kozh, bihan

The 10 most frequent ADJ types: ret, nevez, bras, mat, yaouank, all, gentañ, bihan, kentañ, brav

The 10 most frequent ambiguous lemmas: bras (ADJ 20, ADV 2), mat (ADJ 17, ADV 2), kentañ (ADJ 14, NUM 2), all (ADJ 11, ADV 1), eil (ADJ 5, NUM 4), memes (ADJ 3, X 2), pell (ADJ 3, ADV 1), diavaez (ADJ 2, NOUN 1), gwir (ADJ 2, NOUN 2), marv (NOUN 3, ADJ 2)

The 10 most frequent ambiguous types: mat (ADJ 13, ADV 1), all (ADJ 11, ADV 1), kentañ (ADJ 6, NUM 1), eil (NUM 4, ADJ 3), memes (ADJ 3, X 2), pell (ADJ 3, ADV 1), vat (ADJ 3, ADV 2), brudet (ADJ 2, VERB 1), diavaez (ADJ 2, NOUN 1), marv (ADJ 2, NOUN 2)

• mat
  • ADJ 13: Beajoù a c’hellfe bezañ mat evit ho labour .
  • ADV 1: Evel-se e c’haller stardañ mat ar perlez .
• all
  • ADJ 11: Merket e vo poentoù gane oc’h pa dizhoc’h harnez ar c’hoarierien all .
  • ADV 1: Treuzit an holl wiadoù , tennit ho nadoz er-maez en ur zilec’hiañ anezhi , hag all .
• kentañ
  • ADJ 6: Unan eus ar riezoù kentañ bet krouet en Europa eo ivez .
  • NUM 1: D’ ar sadorn 5 ha Sul 6 a viz Ebrel e vo aozet e Brest ul lodenn eus kampionad Bro-C’hall kentañ rummad mell-droad war gador-ruilh .
• eil
  • NUM 4: Da eil e vez erlec’hiet genoù liesseurt -kenañ gant un doare gen hepken .
  • ADJ 3: Degouezhet eo da eil .
• memes
  • ADJ 3: Katalog menegerioù lodenn ar Rannvro atebek a oa bet kinniget da-geñver ar memes dalc’h .
  • X 2: Hag int kroget da c’hounit tachenn en-dro memes tra .
• pell
  • ADJ 3: Peurrest ar skipailhoù a seblant pell a-walc’h eus al live avat .
  • ADV 1: En em gavout a ra eno gant e gamaladed skol met ivez gant izili eus e familh , deuet eus pell .
• vat
  • ADJ 3: Sin vat eo .
  • ADV 2: Da eil e c’helloc’h stabilaat an traoù da vat ha diskuizhañ un tamm .
• brudet
  • ADJ 2: Ur gevezadeg ziaes ha brudet eo an Olimpiadennoù , met ouzhpenn an dra -se ez int .
  • VERB 1: Kement abadenn sevenadurel a denn da zanvez Breizh a vez brudet gant pedadennoù divyezhek .
• diavaez
  • ADJ 2: istor diavaez ar brezhoneg
  • NOUN 1: Met gant an dispac’h greantel e voe desachet ur bern tud eus an diavaez da Su Kembre .
• marv
  • ADJ 2: Ganet eo bet e 1958 e Naoned ha marv e Pariz e 1993 .
  • NOUN 2: Lakaet e oa izili ar sened d’ ar marv hag ar “ peurrest “ eus ar bobl a oa gwerzhet evel sklaved .

Morphology

The form / lemma ratio of ADJ is 1.140000 (the average of all parts of speech is 1.395664).

The 1st highest number of forms (4) was observed with the lemma “bras”: bras, brasañ, vras, vrasañ.

The 2nd highest number of forms (3) was observed with the lemma “gall”: c’hall, c’hall, gall.

The 3rd highest number of forms (3) was observed with the lemma “hir”: hir, hirañ, hiroc’h.

ADJ occurs with 4 features: Degree (23; 5% instances), NumType (1; 0% instances), Number (1; 0% instances), PronType (1; 0% instances)

ADJ occurs with 5 feature-value pairs: Degree=Cmp, Degree=Sup, NumType=Ord, Number=Sing, PronType=Int

ADJ occurs with 6 feature combinations. The most frequent feature combination is _ (409 tokens). Examples: ret, nevez, bras, mat, yaouank, all, gentañ, bihan, kentañ, brav

Relations

ADJ nodes are attached to their parents using 15 different relations: amod (275; 63% instances), root (77; 18% instances), conj (16; 4% instances), advmod (15; 3% instances), obl (12; 3% instances), advcl (8; 2% instances), acl (6; 1% instances), obj (6; 1% instances), xcomp (6; 1% instances), parataxis (5; 1% instances), ccomp (4; 1% instances), dep (2; 0% instances), compound (1; 0% instances), nmod (1; 0% instances), nsubj (1; 0% instances)

Parents of ADJ nodes belong to 8 different parts of speech: NOUN (276; 63% instances), (77; 18% instances), VERB (57; 13% instances), ADJ (14; 3% instances), PRON (6; 1% instances), PROPN (3; 1% instances), INTJ (1; 0% instances), NUM (1; 0% instances)

273 (63%) ADJ nodes are leaves.

48 (11%) ADJ nodes have one child.

20 (5%) ADJ nodes have two children.

94 (22%) ADJ nodes have three or more children.

The highest child degree of a ADJ node is 7.

Children of ADJ nodes are attached using 23 different relations: cop (96; 20% instances), punct (92; 19% instances), nsubj (56; 12% instances), advmod (47; 10% instances), aux (42; 9% instances), obl (34; 7% instances), conj (19; 4% instances), cc (17; 4% instances), csubj (16; 3% instances), case (12; 2% instances), nmod (11; 2% instances), det (9; 2% instances), mark (8; 2% instances), advcl (6; 1% instances), dep (6; 1% instances), acl (3; 1% instances), amod (2; 0% instances), parataxis (2; 0% instances), dislocated (1; 0% instances), nmod:gen (1; 0% instances), nummod (1; 0% instances), obj (1; 0% instances), xcomp (1; 0% instances)

Children of ADJ nodes belong to 14 different parts of speech: AUX (138; 29% instances), PUNCT (92; 19% instances), NOUN (78; 16% instances), ADV (47; 10% instances), VERB (35; 7% instances), PRON (24; 5% instances), CCONJ (17; 4% instances), ADJ (14; 3% instances), ADP (14; 3% instances), DET (9; 2% instances), SCONJ (6; 1% instances), PROPN (5; 1% instances), NUM (3; 1% instances), PART (1; 0% instances)

© 2014–2024 Universal Dependencies contributors. Site powered by Annodoc and brat

.