home edit page issue tracker

This page pertains to UD version 2.

Treebank Statistics: UD_Serbian-SET: POS Tags: ADJ

There are 2381 ADJ lemmas (23%), 5125 ADJ types (27%) and 11661 ADJ tokens (12%). Out of 17 observed tags, the rank of ADJ is: 2 in number of lemmas, 2 in number of types and 3 in number of tokens.

The 10 most frequent ADJ lemmas: nov, velik, drugi, politički, srpski, međunarodni, prvi, evropski, državni, ekonomski

The 10 most frequent ADJ types: novi, prošle, drugi, inostranih, srpske, bivši, prvi, novog, političke, drugim

The 10 most frequent ambiguous lemmas: velik (ADJ 219, PROPN 1), politički (ADJ 177, ADV 6), ekonomski (ADJ 97, ADV 2), dobar (ADJ 80, ADV 1), važan (ADJ 64, ADV 2), svetski (ADJ 60, ADV 2), demokratski (ADJ 56, ADV 1), potreban (ADJ 51, ADV 1), održati (ADJ 50, VERB 20), bugarski (ADJ 43, PROPN 1)

The 10 most frequent ambiguous types: nove (ADJ 32, NOUN 1), politički (ADJ 25, ADV 5), potrebno (ADJ 19, ADV 1), ekonomski (ADJ 18, ADV 2), moguće (ADJ 15, ADV 4), drugo (ADJ 12, ADV 1), strane (NOUN 59, ADJ 12), sve (DET 70, PART 21, ADJ 11, ADV 6), devedesetih (ADJ 11, NOUN 1), mali (ADJ 9, NOUN 1)

Morphology

The form / lemma ratio of ADJ is 2.152457 (the average of all parts of speech is 1.845258).

The 1st highest number of forms (27) was observed with the lemma “velik”: Veliko, najveća, najveće, najvećeg, najvećem, najveći, najvećih, najvećim, najveću, velika, velike, veliki, velikih, velikim, velikog, velikoj, velikom, veliku, veća, veće, većeg, većem, veći, većih, većim, većoj, veću.

The 2nd highest number of forms (25) was observed with the lemma “dobar”: bolja, bolje, boljeg, boljem, bolji, boljih, bolju, dobar, dobra, dobre, dobri, dobrim, dobro, dobrog, dobrom, dobru, nabolju, najbolja, najbolje, najboljeg, najboljem, najbolji, najboljih, najboljim, najbolju.

The 3rd highest number of forms (20) was observed with the lemma “nov”: najnovija, najnovije, najnovijem, najnoviji, najnovijih, najnovijim, najnoviju, nov, nova, nove, novi, novih, novije, novijoj, novim, novo, novog, novoj, novom, novu.

ADJ occurs with 10 features: Case (10967; 94% instances), Degree (10967; 94% instances), Gender (10967; 94% instances), Number (10967; 94% instances), Definite (10562; 91% instances), VerbForm (1094; 9% instances), Voice (1094; 9% instances), NumType (685; 6% instances), Animacy (496; 4% instances), Poss (137; 1% instances)

ADJ occurs with 22 feature-value pairs: Animacy=Anim, Animacy=Inan, Case=Acc, Case=Dat, Case=Gen, Case=Ins, Case=Loc, Case=Nom, Definite=Def, Definite=Ind, Degree=Cmp, Degree=Pos, Degree=Sup, Gender=Fem, Gender=Masc, Gender=Neut, NumType=Ord, Number=Plur, Number=Sing, Poss=Yes, VerbForm=Part, Voice=Pass

ADJ occurs with 193 feature combinations. The most frequent feature combination is Case=Gen|Definite=Def|Degree=Pos|Gender=Fem|Number=Sing (910 tokens). Examples: prošle, Crne, srpske, evropske, jugoistočne, demokratske, međunarodne, političke, velike, opozicione

Relations

ADJ nodes are attached to their parents using 17 different relations: amod (9223; 79% instances), root (603; 5% instances), acl (507; 4% instances), conj (487; 4% instances), ccomp (184; 2% instances), xcomp (138; 1% instances), nmod (111; 1% instances), obl (102; 1% instances), advcl (87; 1% instances), nsubj (80; 1% instances), flat (48; 0% instances), parataxis (37; 0% instances), csubj (23; 0% instances), appos (11; 0% instances), obj (9; 0% instances), discourse (6; 0% instances), orphan (5; 0% instances)

Parents of ADJ nodes belong to 12 different parts of speech: NOUN (9554; 82% instances), (603; 5% instances), VERB (570; 5% instances), ADJ (470; 4% instances), PROPN (322; 3% instances), ADV (53; 0% instances), DET (26; 0% instances), NUM (22; 0% instances), AUX (21; 0% instances), PRON (17; 0% instances), X (2; 0% instances), PART (1; 0% instances)

9038 (78%) ADJ nodes are leaves.

971 (8%) ADJ nodes have one child.

289 (2%) ADJ nodes have two children.

1363 (12%) ADJ nodes have three or more children.

The highest child degree of a ADJ node is 10.

Children of ADJ nodes are attached using 27 different relations: punct (1375; 17% instances), obl (1372; 17% instances), nsubj (1055; 13% instances), aux (825; 10% instances), cop (659; 8% instances), advmod (531; 7% instances), conj (495; 6% instances), cc (412; 5% instances), mark (360; 5% instances), case (141; 2% instances), parataxis (139; 2% instances), advcl (127; 2% instances), flat (118; 1% instances), discourse (65; 1% instances), ccomp (59; 1% instances), csubj (52; 1% instances), xcomp (38; 0% instances), acl (29; 0% instances), nmod (28; 0% instances), amod (18; 0% instances), det (12; 0% instances), obj (9; 0% instances), nummod:gov (6; 0% instances), appos (5; 0% instances), nummod (3; 0% instances), det:numgov (2; 0% instances), orphan (1; 0% instances)

Children of ADJ nodes belong to 15 different parts of speech: NOUN (1980; 25% instances), AUX (1491; 19% instances), PUNCT (1375; 17% instances), ADV (568; 7% instances), ADJ (470; 6% instances), CCONJ (416; 5% instances), VERB (413; 5% instances), SCONJ (367; 5% instances), PROPN (275; 3% instances), DET (262; 3% instances), ADP (138; 2% instances), PRON (100; 1% instances), PART (45; 1% instances), NUM (25; 0% instances), X (11; 0% instances)