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

Treebank Statistics: UD_Hebrew-HTB: POS Tags: PROPN

There are 2575 PROPN lemmas (23%), 2528 PROPN types (13%) and 7912 PROPN tokens (5%). Out of 15 observed tags, the rank of PROPN is: 2 in number of lemmas, 3 in number of types and 7 in number of tokens.

The 10 most frequent PROPN lemmas: ישראל, ירושלים, כהנא, מכבי, ארה”ב, עיראק, כנסת, צה”ל, יורק, ניו

The 10 most frequent PROPN types: ישראל, ירושלים, כהנא, מכבי, ארה”ב, עיראק, כנסת, צה”ל, ניו, יורק

The 10 most frequent ambiguous lemmas: מכבי (PROPN 95, NOUN 3), חיפה (PROPN 42, VERB 1), כך (PRON 204, ADV 117, PROPN 35), מ (ADP 1688, PROPN 34, NUM 2, X 2, NOUN 1), אל (ADP 146, PROPN 27, NOUN 12, ADJ 2), בן (NOUN 167, PROPN 27), חיים (NOUN 91, PROPN 24), ספטמבר (PROPN 23, NOUN 1), אריה (PROPN 21, NOUN 3), דוד (PROPN 20, NOUN 9)

The 10 most frequent ambiguous types: מכבי (PROPN 95, VERB 2), כנסת (PROPN 59, NOUN 1), ניו (PROPN 49, NOUN 1), גרמניה (PROPN 42, ADJ 1), חיפה (PROPN 42, VERB 1), מ (ADP 1690, PROPN 36, NOUN 2, NUM 2, X 2), כך (PRON 204, ADV 117, PROPN 35), מאיר (PROPN 32, VERB 1), אל (ADP 79, PROPN 27, NOUN 12, ADV 7, ADJ 2), בן (NOUN 73, PROPN 27, ADP 1)

Morphology

The form / lemma ratio of PROPN is 0.981748 (the average of all parts of speech is 1.702584).

The 1st highest number of forms (3) was observed with the lemma “אטאלנטה”: אטאלאנטה, אטאלנטה, אטלנטה.

The 2nd highest number of forms (3) was observed with the lemma “ב”: ב, בי, בך.

The 3rd highest number of forms (3) was observed with the lemma “ואשינגטון”: ואשינגטון, וואשינגטון, וושינגטון.

PROPN occurs with 1 features: Abbr (376; 5% instances)

PROPN occurs with 1 feature-value pairs: Abbr=Yes

PROPN occurs with 2 feature combinations. The most frequent feature combination is _ (7536 tokens). Examples: ישראל, ירושלים, כהנא, מכבי, עיראק, כנסת, ניו, יורק, גרמניה, חיפה

Relations

PROPN nodes are attached to their parents using 18 different relations: flat:name (2344; 30% instances), nsubj (1111; 14% instances), nmod (989; 13% instances), compound:smixut (898; 11% instances), obl (767; 10% instances), conj (427; 5% instances), appos (401; 5% instances), nmod:poss (341; 4% instances), dep (275; 3% instances), obj (175; 2% instances), root (96; 1% instances), nsubj:cop (40; 1% instances), acl:relcl (23; 0% instances), parataxis (12; 0% instances), ccomp (6; 0% instances), advcl (3; 0% instances), acl (2; 0% instances), amod (2; 0% instances)

Parents of PROPN nodes belong to 14 different parts of speech: NOUN (3680; 47% instances), PROPN (1982; 25% instances), VERB (1901; 24% instances), ADJ (98; 1% instances), (96; 1% instances), NUM (56; 1% instances), AUX (26; 0% instances), DET (21; 0% instances), PRON (19; 0% instances), ADV (16; 0% instances), X (10; 0% instances), CCONJ (4; 0% instances), ADP (2; 0% instances), SCONJ (1; 0% instances)

3824 (48%) PROPN nodes are leaves.

2385 (30%) PROPN nodes have one child.

899 (11%) PROPN nodes have two children.

804 (10%) PROPN nodes have three or more children.

The highest child degree of a PROPN node is 22.

Children of PROPN nodes are attached using 26 different relations: case (1664; 22% instances), flat:name (1527; 20% instances), punct (1486; 20% instances), conj (491; 6% instances), case:gen (473; 6% instances), nmod (253; 3% instances), cc (248; 3% instances), appos (224; 3% instances), dep (214; 3% instances), det (179; 2% instances), acl:relcl (176; 2% instances), case:acc (171; 2% instances), nummod (136; 2% instances), advmod (67; 1% instances), amod (66; 1% instances), nsubj (50; 1% instances), cop (35; 0% instances), mark (31; 0% instances), acl (28; 0% instances), nsubj:cop (24; 0% instances), nmod:poss (18; 0% instances), parataxis (18; 0% instances), advcl (7; 0% instances), ccomp (2; 0% instances), xcomp (2; 0% instances), obl (1; 0% instances)

Children of PROPN nodes belong to 14 different parts of speech: ADP (2303; 30% instances), PROPN (1982; 26% instances), PUNCT (1486; 20% instances), NOUN (698; 9% instances), CCONJ (263; 3% instances), VERB (208; 3% instances), NUM (185; 2% instances), DET (180; 2% instances), ADJ (100; 1% instances), ADV (73; 1% instances), PRON (65; 1% instances), SCONJ (36; 0% instances), AUX (11; 0% instances), X (1; 0% instances)