home edit page issue tracker
This page pertains to UD version 2.

Treebank Statistics: UD_Armenian-ArmTDP: POS Tags: PROPN

There are 1721 PROPN lemmas (14%), 2427 PROPN types (10%) and 5309 PROPN tokens (5%). Out of 17 observed tags, the rank of PROPN is: 3 in number of lemmas, 3 in number of types and 6 in number of tokens.

The 10 most frequent PROPN lemmas: Հայաստան, Երևան, ՀՀ, ՍՍՀՄ, Ա., ՍՄԿԿ, Վ., Իրան, Սարգսյան, Մ.

The 10 most frequent PROPN types: Հայաստանի, ՀՀ, ՍՍՀՄ, Ա, Երևանի, ՍՄԿԿ, Վ, Հայաստանում, Մ, Լևոն

The 10 most frequent ambiguous lemmas: Գ. (PROPN 29, NOUN 1), Հ. (PROPN 19, ADV 1), Սունդուկյան (PROPN 19, NOUN 3), Խաչատրյան (PROPN 9, NOUN 1), Դիլիջան (PROPN 7, NOUN 4), Անգլիա (PROPN 5, NOUN 1), Ամերիկա (NOUN 4, PROPN 4), Գնդեվանք (PROPN 3, NOUN 1), Լենինգրադ (PROPN 3, NOUN 2), Ղազախստան (PROPN 3, NOUN 2)

The 10 most frequent ambiguous types: Ա (PROPN 56, ADJ 8, X 1), Մ (PROPN 34, X 1), Գ (PROPN 29, NOUN 1, X 1), Լ (PROPN 25, ADJ 3), Ս (PROPN 25, ADJ 9), Հ (PROPN 19, ADV 1), Ի (ADP 19, PROPN 17, NOUN 3), Ն (PROPN 9, ADJ 8, DET 2), Պ (PROPN 9, VERB 1), Դ (PROPN 7, NOUN 1)

Morphology

The form / lemma ratio of PROPN is 1.410227 (the average of all parts of speech is 1.883575).

The 1st highest number of forms (9) was observed with the lemma “Հայաստան”: ՀԱՅԱՍՏԱՆ, ՀԱՅԱՍՏԱՆԻ, Հայաստան, Հայաստանը, Հայաստանի, Հայաստանին, Հայաստանից, Հայաստանն, Հայաստանում.

The 2nd highest number of forms (7) was observed with the lemma “Ադրբեջան”: Ադրբեջան, Ադրբեջանը, Ադրբեջանի, Ադրբեջանին, Ադրբեջանից, Ադրբեջանով, Ադրբեջանում.

The 3rd highest number of forms (7) was observed with the lemma “Երևան”: ԵՐԵՎԱՆ, ԵՐԵՎԱՆԻ, Երևան, Երևանը, Երևանի, Երևանից, Երևանում.

PROPN occurs with 9 features: Animacy (5309; 100% instances), Case (5309; 100% instances), Definite (5308; 100% instances), Number (5308; 100% instances), NameType (5301; 100% instances), Abbr (848; 16% instances), Style (37; 1% instances), Number[psor] (1; 0% instances), Typo (1; 0% instances)

PROPN occurs with 31 feature-value pairs: Abbr=Yes, Animacy=Hum, Animacy=Nhum, Case=Abl, Case=Dat, Case=Gen, Case=Ins, Case=Loc, Case=Nom, Definite=Def, Definite=Ind, NameType=Com, NameType=Geo, NameType=Giv, NameType=Oth, NameType=Pat, NameType=Pro, NameType=Prs, NameType=Sur, Number=Coll, Number=Plur, Number=Ptan, Number=Sing, Number[psor]=Sing, Style=Arch, Style=Coll, Style=Expr, Style=Rare, Style=Slng, Style=Vrnc, Typo=Yes

PROPN occurs with 87 feature combinations. The most frequent feature combination is Animacy=Nhum|Case=Dat|Definite=Ind|NameType=Geo|Number=Sing (922 tokens). Examples: Հայաստանի, Երևանի, Իրանի, Ադրբեջանի, Եվրոպայի, Սևանի, Մեղրու, Ֆրանսիայի, Էջմիածնի, Լենինականի

Relations

PROPN nodes are attached to their parents using 30 different relations: nmod:poss (1235; 23% instances), flat:name (1092; 21% instances), nsubj (696; 13% instances), conj (610; 11% instances), nmod (567; 11% instances), obl (327; 6% instances), root (173; 3% instances), appos (116; 2% instances), nmod:npmod (96; 2% instances), parataxis (80; 2% instances), obj (73; 1% instances), iobj (58; 1% instances), list (44; 1% instances), flat (42; 1% instances), nsubj:pass (17; 0% instances), orphan (14; 0% instances), vocative (14; 0% instances), flat:range (12; 0% instances), xcomp (11; 0% instances), compound (8; 0% instances), dislocated (5; 0% instances), acl:relcl (4; 0% instances), advcl (4; 0% instances), ccomp (3; 0% instances), csubj (2; 0% instances), obl:agent (2; 0% instances), acl (1; 0% instances), compound:redup (1; 0% instances), dep (1; 0% instances), nsubj:caus (1; 0% instances)

Parents of PROPN nodes belong to 13 different parts of speech: NOUN (2172; 41% instances), PROPN (1730; 33% instances), VERB (1123; 21% instances), (173; 3% instances), ADJ (38; 1% instances), NUM (24; 0% instances), PRON (22; 0% instances), X (16; 0% instances), DET (5; 0% instances), ADV (3; 0% instances), AUX (1; 0% instances), PART (1; 0% instances), SYM (1; 0% instances)

2739 (52%) PROPN nodes are leaves.

1069 (20%) PROPN nodes have one child.

771 (15%) PROPN nodes have two children.

730 (14%) PROPN nodes have three or more children.

The highest child degree of a PROPN node is 11.

Children of PROPN nodes are attached using 36 different relations: punct (1904; 35% instances), flat:name (1110; 21% instances), conj (624; 12% instances), nmod (487; 9% instances), cc (211; 4% instances), amod (191; 4% instances), dep (135; 3% instances), case (118; 2% instances), flat (88; 2% instances), parataxis (84; 2% instances), acl:relcl (55; 1% instances), advmod:emph (48; 1% instances), appos (48; 1% instances), cop (41; 1% instances), list (41; 1% instances), acl (36; 1% instances), nsubj (31; 1% instances), nmod:poss (26; 0% instances), orphan (24; 0% instances), nmod:npmod (20; 0% instances), det (12; 0% instances), flat:range (12; 0% instances), compound (11; 0% instances), det:poss (7; 0% instances), discourse (6; 0% instances), advcl (5; 0% instances), mark (5; 0% instances), obl (5; 0% instances), nummod (3; 0% instances), advmod (2; 0% instances), aux (2; 0% instances), case:loc (2; 0% instances), compound:redup (1; 0% instances), dislocated (1; 0% instances), expl (1; 0% instances), xcomp (1; 0% instances)

Children of PROPN nodes belong to 16 different parts of speech: PUNCT (1904; 35% instances), PROPN (1730; 32% instances), NOUN (880; 16% instances), CCONJ (208; 4% instances), ADJ (197; 4% instances), ADP (114; 2% instances), VERB (102; 2% instances), NUM (83; 2% instances), AUX (43; 1% instances), ADV (42; 1% instances), PRON (35; 1% instances), DET (26; 0% instances), X (16; 0% instances), PART (11; 0% instances), SCONJ (4; 0% instances), SYM (3; 0% instances)