home edit page issue tracker

This page pertains to UD version 2.

Treebank Statistics: UD_Croatian: POS Tags: PROPN

There are 4511 PROPN lemmas (22%), 5578 PROPN types (15%) and 12691 PROPN tokens (6%). Out of 17 observed tags, the rank of PROPN is: 2 in number of lemmas, 3 in number of types and 6 in number of tokens.

The 10 most frequent PROPN lemmas: Hrvatska, EU, Kosovo, Srbija, BiH, Makedonija, Turska, Europa, Bugarska, NATO

The 10 most frequent PROPN types: EU, BiH, Kosova, Hrvatskoj, Srbije, SETimes, Hrvatska, Kosovo, Srbija, Hrvatske

The 10 most frequent ambiguous lemmas: Hrvatska (PROPN 307, ADJ 2), BiH (PROPN 157, NOUN 1), SETimes (PROPN 104, X 6), Gora (PROPN 91, NOUN 1), Crna (PROPN 90, NOUN 1), Windows (PROPN 74, NOUN 2), Vlada (PROPN 39, NOUN 1), Ivan (PROPN 19, AUX 1), Internet (PROPN 18, ADJ 1, NOUN 1), Bog (NOUN 25, PROPN 17)

The 10 most frequent ambiguous types: BiH (PROPN 157, AUX 4, NOUN 1), SETimes (PROPN 102, X 5), Hrvatska (PROPN 96, ADJ 7), Hrvatske (PROPN 66, ADJ 19), Europe (PROPN 56, X 1), Bugarska (PROPN 44, ADJ 2), Turske (PROPN 43, ADJ 2), Turska (PROPN 40, ADJ 1), Crne (PROPN 33, ADJ 3), Bugarske (PROPN 28, ADJ 1)

Morphology

The form / lemma ratio of PROPN is 1.236533 (the average of all parts of speech is 1.827681).

The 1st highest number of forms (6) was observed with the lemma “Gora”: Gora, Gore, Gori, Gorin, Gorom, Goru.

The 2nd highest number of forms (6) was observed with the lemma “Hercegovina”: Hercegovina, Hercegovine, Hercegovini, Hercegovinom, Hercegovinu, Hercegovni.

The 3rd highest number of forms (6) was observed with the lemma “Hrvatska”: HRVATSKE, Hrvatska, Hrvatske, Hrvatskoj, Hrvatskom, Hrvatsku.

PROPN occurs with 5 features: Case (12691; 100% instances), Gender (12691; 100% instances), Number (12691; 100% instances), Animacy (593; 5% instances), Polarity (3; 0% instances)

PROPN occurs with 15 feature-value pairs: Animacy=Anim, Animacy=Inan, Case=Acc, Case=Dat, Case=Gen, Case=Ins, Case=Loc, Case=Nom, Case=Voc, Gender=Fem, Gender=Masc, Gender=Neut, Number=Plur, Number=Sing, Polarity=Neg

PROPN occurs with 35 feature combinations. The most frequent feature combination is Case=Nom|Gender=Masc|Number=Sing (4581 tokens). Examples: Windows, Sanader, Zagreb, Crvenkovski, Erdogan, May, Nikola, HDZ, Babić, Rehn

Relations

PROPN nodes are attached to their parents using 27 different relations: nmod (2885; 23% instances), nsubj (2311; 18% instances), flat (2130; 17% instances), appos (1839; 14% instances), conj (1142; 9% instances), obl (1009; 8% instances), compound (518; 4% instances), obj (328; 3% instances), parataxis (199; 2% instances), nsubj:pass (73; 1% instances), amod (64; 1% instances), iobj (52; 0% instances), root (51; 0% instances), flat:foreign (12; 0% instances), xcomp (12; 0% instances), orphan (10; 0% instances), acl (9; 0% instances), det (9; 0% instances), nummod (9; 0% instances), advmod (7; 0% instances), ccomp (5; 0% instances), discourse (4; 0% instances), vocative (4; 0% instances), advcl (3; 0% instances), list (3; 0% instances), case (2; 0% instances), aux (1; 0% instances)

Parents of PROPN nodes belong to 14 different parts of speech: NOUN (5174; 41% instances), PROPN (3540; 28% instances), VERB (3150; 25% instances), ADJ (534; 4% instances), ADV (77; 1% instances), (51; 0% instances), AUX (50; 0% instances), ADP (34; 0% instances), X (28; 0% instances), PRON (20; 0% instances), DET (16; 0% instances), NUM (11; 0% instances), SYM (4; 0% instances), PART (2; 0% instances)

6126 (48%) PROPN nodes are leaves.

4385 (35%) PROPN nodes have one child.

1441 (11%) PROPN nodes have two children.

739 (6%) PROPN nodes have three or more children.

The highest child degree of a PROPN node is 13.

Children of PROPN nodes are attached using 33 different relations: case (2362; 23% instances), flat (2144; 21% instances), punct (1467; 14% instances), conj (1174; 12% instances), cc (671; 7% instances), compound (477; 5% instances), amod (404; 4% instances), appos (356; 4% instances), nmod (289; 3% instances), acl (204; 2% instances), parataxis (90; 1% instances), discourse (84; 1% instances), nummod (80; 1% instances), advmod (77; 1% instances), cop (53; 1% instances), nsubj (47; 0% instances), det (44; 0% instances), orphan (25; 0% instances), aux (21; 0% instances), mark (15; 0% instances), flat:foreign (13; 0% instances), obj (6; 0% instances), advcl (3; 0% instances), list (3; 0% instances), xcomp (3; 0% instances), aux:pass (2; 0% instances), ccomp (2; 0% instances), dislocated (2; 0% instances), goeswith (2; 0% instances), nsubj:pass (2; 0% instances), csubj (1; 0% instances), dep (1; 0% instances), expl (1; 0% instances)

Children of PROPN nodes belong to 16 different parts of speech: PROPN (3540; 35% instances), ADP (2400; 24% instances), PUNCT (1467; 14% instances), CCONJ (776; 8% instances), NOUN (764; 8% instances), ADJ (535; 5% instances), VERB (153; 2% instances), NUM (127; 1% instances), ADV (111; 1% instances), AUX (76; 1% instances), DET (43; 0% instances), X (43; 0% instances), PART (36; 0% instances), SCONJ (36; 0% instances), PRON (17; 0% instances), SYM (1; 0% instances)