home edit page issue tracker

This page pertains to UD version 2.

Treebank Statistics: UD_Maghrebi_Arabic_French-Arabizi: POS Tags: SCONJ

There are 27 SCONJ lemmas (0%), 78 SCONJ types (1%) and 230 SCONJ tokens (1%). Out of 16 observed tags, the rank of SCONJ is: 13 in number of lemmas, 10 in number of types and 13 in number of tokens.

The 10 most frequent SCONJ lemmas: que, si, comme, parce, parce_que, puisque, qu’, qu’eux, quand, sinon

The 10 most frequent SCONJ types: que, ida, kima, loukan, si, puisque, parce, ana, beli, bli

The 10 most frequent ambiguous lemmas: que (SCONJ 90, PART 60, ADV 14, ADP 12, PRON 12, X 1), si (SCONJ 58, INTJ 2, CCONJ 1), comme (ADP 62, SCONJ 17, ADV 11, PRON 1), parce_que (SCONJ 11, ADP 1, ADV 1, CCONJ 1), qu’ (SCONJ 6, PRON 2), quand (ADV 33, ADP 2, SCONJ 2, PRON 1), sinon (ADV 2, CCONJ 2, PART 2, SCONJ 2), tant_que (SCONJ 2, ADV 1), comment (ADV 19, SCONJ 1), contre (ADP 23, SCONJ 1)

The 10 most frequent ambiguous types: que (SCONJ 62, PRON 5, ADV 1), kima (ADP 27, SCONJ 7, ADV 2, NOUN 1), si (NOUN 9, SCONJ 8, VERB 7, CCONJ 2, INTJ 2, ADV 1, PRON 1), ana (PRON 50, SCONJ 4, ADP 1, X 1), ina (VERB 6, SCONJ 4, PRON 1, X 1), ke (SCONJ 4, ADP 1), ki (ADV 34, ADP 15, PRON 4, SCONJ 4), loukane (SCONJ 4, VERB 2), par (ADP 6, SCONJ 4), in (INTJ 6, SCONJ 3)

Morphology

The form / lemma ratio of SCONJ is 2.888889 (the average of all parts of speech is 1.474223).

The 1st highest number of forms (23) was observed with the lemma “si”: an, c, ida, ila, in, inna, kon, koun, koune, la, lakane, laou, lokan, lokane, lou, loukan, loukane, louken, lw, si, ya, yaloukan, yla.

The 2nd highest number of forms (13) was observed with the lemma “que”: an, ana, balli, beli, belli, bili, bli, ina, ke, man, que, quille, wue.

The 3rd highest number of forms (9) was observed with the lemma “comme”: chghoul, com, comme, ki, kif, kih, kima, kimha, kom.

SCONJ occurs with 2 features: Typo (8; 3% instances), AdpType (6; 3% instances)

SCONJ occurs with 2 feature-value pairs: AdpType=Prep, Typo=Yes

SCONJ occurs with 3 feature combinations. The most frequent feature combination is _ (216 tokens). Examples: que, ida, kima, loukan, si, puisque, parce, ana, ina, ke

Relations

SCONJ nodes are attached to their parents using 7 different relations: mark (201; 87% instances), fixed (14; 6% instances), case (8; 3% instances), dep (2; 1% instances), discourse (2; 1% instances), nsubj (2; 1% instances), obl (1; 0% instances)

Parents of SCONJ nodes belong to 10 different parts of speech: VERB (146; 63% instances), NOUN (30; 13% instances), PROPN (18; 8% instances), ADJ (13; 6% instances), SCONJ (11; 5% instances), PRON (6; 3% instances), ADV (2; 1% instances), AUX (2; 1% instances), CCONJ (1; 0% instances), NUM (1; 0% instances)

211 (92%) SCONJ nodes are leaves.

14 (6%) SCONJ nodes have one child.

5 (2%) SCONJ nodes have two children.

The highest child degree of a SCONJ node is 2.

Children of SCONJ nodes are attached using 6 different relations: fixed (14; 58% instances), goeswith (6; 25% instances), amod (1; 4% instances), case (1; 4% instances), cc (1; 4% instances), punct (1; 4% instances)

Children of SCONJ nodes belong to 8 different parts of speech: SCONJ (11; 46% instances), X (6; 25% instances), ADP (2; 8% instances), ADJ (1; 4% instances), ADV (1; 4% instances), CCONJ (1; 4% instances), PRON (1; 4% instances), PUNCT (1; 4% instances)