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

Treebank Statistics: UD_Zaar-Autogramm: POS Tags: PART

There are 34 PART lemmas (2%), 91 PART types (3%) and 1807 PART tokens (9%). Out of 16 observed tags, the rank of PART is: 11 in number of lemmas, 8 in number of types and 5 in number of tokens.

The 10 most frequent PART lemmas: tòː, hŋ́, ɗi, bàː, ŋaː, kən, oː, ni, kúmá, mə́n

The 10 most frequent PART types: tôː, hŋ́, ɗi, bàː, tòː, oː, ni, aː, kúmá, máː

The 10 most frequent ambiguous lemmas: tòː (PART 442, INTJ 17, CCONJ 1), ŋaː (PART 118, ADJ 46, NOUN 12, ADV 2), ni (PART 94, X 1), kúmá (PART 67, CCONJ 6, X 1), mə́n (PART 64, NOUN 44), máː (PART 62, ADV 28, X 1), ɗa (ADP 56, ADV 35, PART 34, SCONJ 18), eː (PART 22, INTJ 5), kóː (CCONJ 92, SCONJ 12, PART 11, DET 5, ADV 3, ADP 1, X 1), bâː (PART 10, INTJ 1)

The 10 most frequent ambiguous types: tôː (PART 327, INTJ 12, CCONJ 1), tòː (PART 113, INTJ 5), ni (PART 85, X 1), aː (PART 81, INTJ 2), kúmá (PART 66, CCONJ 6), máː (PART 62, ADV 28, AUX 24, X 1), mə́n (PART 57, NOUN 55), ɗa (ADP 55, ADV 36, PART 32, SCONJ 29), ɣən (PART 31, AUX 17), ɣəndí (PART 24, AUX 8)

Morphology

The form / lemma ratio of PART is 2.676471 (the average of all parts of speech is 1.729120).

The 1st highest number of forms (15) was observed with the lemma “hŋ́”: hŋ́, hŋ́ə́y, hḿ, n, ń, ň, ŋ, ŋə́y, ŋ́, ŋ̌, ǐn, ə́ŋ, ə̌n, ə̌ŋ, ə̌ːníː.

The 2nd highest number of forms (15) was observed with the lemma “kən”: gən, gəndí, gəní, gə̂n, kən, kəndá, kəndí, kənín, ɣən, ɣəndá, ɣəndí, ɣəní, ɣənín, ɣəŋ, ɣə̂n.

The 3rd highest number of forms (8) was observed with the lemma “ŋaː”: aː, yaː, àː, âː, ŋaː, ŋâː, ŋǎː, ŋǎːŋ.

PART occurs with 6 features: PartType (1806; 100% instances), Polarity (462; 26% instances), Mood (214; 12% instances), Aspect (94; 5% instances), Deixis (22; 1% instances), Foreign (11; 1% instances)

PART occurs with 15 feature-value pairs: Aspect=Iter, Deixis=Prox, Deixis=Remt, Foreign=Yes, Mood=Ast, Mood=Int, Mood=Irr, PartType=Adv, PartType=Disc, PartType=Foc, PartType=Illoc, PartType=Neg, PartType=Pred, PartType=Top, Polarity=Neg

PART occurs with 25 feature combinations. The most frequent feature combination is PartType=Foc (568 tokens). Examples: tôː, tòː, kúmá, ɣən, ɣəndí, kàm, kəndí, wàːtòː, gəndí, kən

Relations

PART nodes are attached to their parents using 9 different relations: discourse (776; 43% instances), advmod (640; 35% instances), compound:prt (355; 20% instances), root (16; 1% instances), fixed (7; 0% instances), advcl (4; 0% instances), ccomp (4; 0% instances), parataxis (4; 0% instances), conj (1; 0% instances)

Parents of PART nodes belong to 16 different parts of speech: VERB (1380; 76% instances), NOUN (178; 10% instances), PRON (89; 5% instances), ADV (40; 2% instances), AUX (27; 1% instances), PROPN (21; 1% instances), INTJ (19; 1% instances), (16; 1% instances), PART (10; 1% instances), SCONJ (8; 0% instances), ADP (6; 0% instances), ADJ (3; 0% instances), DET (3; 0% instances), NUM (3; 0% instances), X (3; 0% instances), CCONJ (1; 0% instances)

1672 (93%) PART nodes are leaves.

101 (6%) PART nodes have one child.

16 (1%) PART nodes have two children.

18 (1%) PART nodes have three or more children.

The highest child degree of a PART node is 6.

Children of PART nodes are attached using 10 different relations: punct (121; 58% instances), nsubj (29; 14% instances), ccomp (14; 7% instances), discourse (11; 5% instances), mark (8; 4% instances), advcl (7; 3% instances), advmod (7; 3% instances), obl (4; 2% instances), xcomp (4; 2% instances), dislocated (2; 1% instances)

Children of PART nodes belong to 10 different parts of speech: PUNCT (121; 58% instances), NOUN (33; 16% instances), VERB (20; 10% instances), PART (10; 5% instances), SCONJ (8; 4% instances), INTJ (4; 2% instances), PRON (4; 2% instances), PROPN (4; 2% instances), ADV (2; 1% instances), CCONJ (1; 0% instances)