home edit page issue tracker

This page pertains to UD version 2.

Treebank Statistics: UD_Polish-PDB: POS Tags: ADJ

There are 6944 ADJ lemmas (22%), 15533 ADJ types (25%) and 35928 ADJ tokens (10%). Out of 17 observed tags, the rank of ADJ is: 2 in number of lemmas, 2 in number of types and 5 in number of tokens.

The 10 most frequent ADJ lemmas: jeden, inny, sam, nowy, duży, pierwszy, cały, drugi, europejski, dobry

The 10 most frequent ADJ types: innych, jeden, sam, inne, europejskiej, pierwszy, różnych, jednym, cały, 1

The 10 most frequent ambiguous lemmas: jeden (ADJ 549, NOUN 2), inny (ADJ 544, NOUN 31), nowy (ADJ 356, NOUN 1), drugi (ADJ 323, NOUN 1), dobry (ADJ 294, NOUN 1), mały (ADJ 283, NOUN 1), młody (ADJ 213, NOUN 3), biały (ADJ 208, NOUN 2), czarny (ADJ 194, NOUN 3), ostatni (ADJ 189, NOUN 1)

The 10 most frequent ambiguous types: innych (ADJ 173, NOUN 17), jednym (ADJ 78, NOUN 1), 1 (X 85, ADJ 77, NUM 3), drugi (ADJ 65, NOUN 1), drugą (ADJ 40, NOUN 3), młodych (ADJ 41, NOUN 2), innymi (ADJ 37, NOUN 1), małe (ADJ 27, NOUN 1), dobry (ADJ 31, NOUN 1), drugie (ADJ 35, NOUN 1)

Morphology

The form / lemma ratio of ADJ is 2.236895 (the average of all parts of speech is 1.966055).

The 1st highest number of forms (30) was observed with the lemma “dobry”: dobra, dobre, dobrego, dobrej, dobry, dobrych, dobrym, dobrymi, dobrzy, dobrą, lepsi, lepsza, lepsze, lepszego, lepszej, lepszemu, lepszy, lepszych, lepszym, lepszą, najlepsi, najlepsza, najlepsze, najlepszego, najlepszej, najlepszy, najlepszych, najlepszym, najlepszymi, najlepszą.

The 2nd highest number of forms (27) was observed with the lemma “duży”: duża, duże, dużego, dużej, duży, dużych, dużym, dużymi, dużą, najwięksi, największa, największe, największego, największej, największy, największych, największym, największą, większa, większe, większego, większej, większemu, większy, większych, większym, większą.

The 3rd highest number of forms (27) was observed with the lemma “ważny”: Ważniejszy, najważniejsi, najważniejsza, najważniejsze, najważniejszej, najważniejszy, najważniejszych, najważniejszym, najważniejszą, ważna, ważne, ważnego, ważnej, ważni, ważniejsi, ważniejsza, ważniejsze, ważniejszego, ważniejszej, ważniejszych, ważniejszym, ważniejszą, ważny, ważnych, ważnym, ważnymi, ważną.

ADJ occurs with 15 features: Case (35595; 99% instances), Gender (35432; 99% instances), Number (35432; 99% instances), Degree (28668; 80% instances), Animacy (16206; 45% instances), Aspect (6764; 19% instances), Polarity (6764; 19% instances), VerbForm (6764; 19% instances), Voice (6764; 19% instances), NumForm (1159; 3% instances), NumType (1159; 3% instances), Hyph (259; 1% instances), PrepCase (163; 0% instances), Abbr (61; 0% instances), Variant (13; 0% instances)

ADJ occurs with 32 feature-value pairs: Abbr=Yes, Animacy=Hum, Animacy=Inan, Animacy=Nhum, Aspect=Imp, Aspect=Perf, Case=Acc, Case=Dat, Case=Gen, Case=Ins, Case=Loc, Case=Nom, Case=Voc, Degree=Cmp, Degree=Pos, Degree=Sup, Gender=Fem, Gender=Masc, Gender=Neut, Hyph=Yes, NumForm=Digit, NumForm=Roman, NumType=Ord, Number=Plur, Number=Sing, Polarity=Neg, Polarity=Pos, PrepCase=Pre, Variant=Short, VerbForm=Part, Voice=Act, Voice=Pass

ADJ occurs with 403 feature combinations. The most frequent feature combination is Case=Gen|Degree=Pos|Gender=Fem|Number=Sing (2271 tokens). Examples: europejskiej, drugiej, polskiej, jednej, nowej, pierwszej, niniejszej, całej, społecznej, publicznej

Relations

ADJ nodes are attached to their parents using 35 different relations: amod (22640; 63% instances), acl (4733; 13% instances), conj (2090; 6% instances), amod:flat (1777; 5% instances), root (1451; 4% instances), obl (672; 2% instances), xcomp (327; 1% instances), nsubj (264; 1% instances), ccomp (255; 1% instances), acl:relcl (224; 1% instances), advcl (199; 1% instances), fixed (161; 0% instances), parataxis:obj (159; 0% instances), obl:arg (149; 0% instances), xcomp:pred (143; 0% instances), nmod (128; 0% instances), ccomp:obj (118; 0% instances), obj (75; 0% instances), parataxis:insert (64; 0% instances), obl:cmpr (60; 0% instances), iobj (58; 0% instances), nmod:arg (58; 0% instances), ccomp:cleft (25; 0% instances), csubj (24; 0% instances), appos (19; 0% instances), orphan (11; 0% instances), advcl:cmpr (8; 0% instances), advcl:relcl (7; 0% instances), nmod:poss (7; 0% instances), obl:agent (6; 0% instances), nmod:pred (5; 0% instances), advmod:emph (4; 0% instances), nsubj:pass (3; 0% instances), vocative (3; 0% instances), csubj:pass (1; 0% instances)

Parents of ADJ nodes belong to 14 different parts of speech: NOUN (28667; 80% instances), VERB (2406; 7% instances), ADJ (2048; 6% instances), (1451; 4% instances), PROPN (804; 2% instances), PRON (250; 1% instances), DET (170; 0% instances), ADV (93; 0% instances), NUM (18; 0% instances), ADP (10; 0% instances), PART (5; 0% instances), X (4; 0% instances), INTJ (1; 0% instances), PUNCT (1; 0% instances)

23828 (66%) ADJ nodes are leaves.

6655 (19%) ADJ nodes have one child.

1878 (5%) ADJ nodes have two children.

3567 (10%) ADJ nodes have three or more children.

The highest child degree of a ADJ node is 12.

Children of ADJ nodes are attached using 54 different relations: punct (5193; 20% instances), obl (3183; 12% instances), conj (2205; 9% instances), advmod (2112; 8% instances), cop (1497; 6% instances), aux:pass (1401; 5% instances), obl:arg (1322; 5% instances), nsubj (1125; 4% instances), cc (1076; 4% instances), nsubj:pass (1047; 4% instances), iobj (801; 3% instances), mark (797; 3% instances), advmod:emph (739; 3% instances), case (693; 3% instances), obl:agent (512; 2% instances), obj (432; 2% instances), advcl (282; 1% instances), nmod:flat (230; 1% instances), expl:pv (226; 1% instances), obl:cmpr (214; 1% instances), parataxis:insert (96; 0% instances), ccomp (67; 0% instances), xcomp (60; 0% instances), det (56; 0% instances), advmod:neg (55; 0% instances), aux:clitic (51; 0% instances), aux:cnd (44; 0% instances), aux (39; 0% instances), xcomp:pred (37; 0% instances), amod (36; 0% instances), acl:relcl (33; 0% instances), csubj (29; 0% instances), cc:preconj (28; 0% instances), nmod (26; 0% instances), list (17; 0% instances), acl (16; 0% instances), advcl:cmpr (14; 0% instances), fixed (12; 0% instances), parataxis:obj (11; 0% instances), ccomp:obj (10; 0% instances), discourse:intj (10; 0% instances), nummod (10; 0% instances), nummod:gov (10; 0% instances), vocative (10; 0% instances), appos (8; 0% instances), amod:flat (7; 0% instances), det:numgov (7; 0% instances), orphan (7; 0% instances), flat (6; 0% instances), csubj:pass (5; 0% instances), xcomp:subj (4; 0% instances), det:nummod (3; 0% instances), det:poss (2; 0% instances), aux:imp (1; 0% instances)

Children of ADJ nodes belong to 16 different parts of speech: NOUN (7636; 29% instances), PUNCT (5193; 20% instances), AUX (3033; 12% instances), ADV (2111; 8% instances), ADJ (2048; 8% instances), CCONJ (1084; 4% instances), PRON (888; 3% instances), VERB (801; 3% instances), PART (792; 3% instances), SCONJ (777; 3% instances), ADP (766; 3% instances), PROPN (445; 2% instances), DET (263; 1% instances), X (34; 0% instances), NUM (33; 0% instances), INTJ (10; 0% instances)