Treebank Statistics: UD_Croatian-SET: Features: PronType
This feature is universal.
It occurs with 7 different values: Dem, Ind, Int, Neg, Prs, Rel, Tot.
Some words have combined values of the feature; 1 combinations have been observed: Int|Rel.
14258 tokens (7%) have a non-empty value of PronType.
521 types (1%) occur at least once with a non-empty value of PronType.
96 lemmas (1%) occur at least once with a non-empty value of PronType.
The feature is used with 3 part-of-speech tags: DET (7495; 4% instances), PRON (5300; 3% instances), ADV (1463; 1% instances).
DET
7495 DET tokens (98% of all DET tokens) have a non-empty value of PronType.
The most frequent other feature values with which DET and PronType co-occurred: Number[psor]=EMPTY (6666; 89%), Person=EMPTY (6666; 89%), Poss=EMPTY (5913; 79%), Number=Sing (5100; 68%).
DET tokens may have the following values of PronType:
Dem(2478; 33% of non-emptyPronType): to, ove, toga, tome, taj, te, ovog, tom, ovaj, ovoInd(441; 6% of non-emptyPronType): nekoliko, neki, neke, nekim, neka, nekog, nekih, neku, neko, nekogaInt,Rel(2636; 35% of non-emptyPronType): koji, koje, koja, kojima, koju, kojoj, kojem, kojim, kojeg, kojihNeg(23; 0% of non-emptyPronType): nikakve, nikakvu, nikakvih, nikakvo, nikakvi, Nikakva, nikakav, ničijuPrs(1499; 20% of non-emptyPronType): svoje, svoj, svoju, svojim, njegova, svojih, njegov, njihov, naše, svojojTot(418; 6% of non-emptyPronType): sve, svaki, svih, svaka, svakog, svega, svake, svi, svakom, svimEMPTY(189): više, mnogo, pola, puno, manje, malo, dosta, dovoljno, najviše, previše
PronType seems to be lexical feature of DET. 100% lemmas (38) occur only with one value of PronType.
PRON
5300 PRON tokens (100% of all PRON tokens) have a non-empty value of PronType.
The most frequent other feature values with which PRON and PronType co-occurred: Gender=EMPTY (3785; 71%), Person=EMPTY (3712; 70%), Number=EMPTY (3711; 70%), Case=Acc (3503; 66%), Reflex=Yes (2899; 55%).
PRON tokens may have the following values of PronType:
Ind(78; 1% of non-emptyPronType): nešto, netko, itko, išta, nečemu, nekome, nečeg, nečega, nekog, nečemInt,Rel(622; 12% of non-emptyPronType): što, tko, čime, čega, čemu, koga, kim, kome, kime, čimNeg(100; 2% of non-emptyPronType): ništa, nitko, nikome, nikoga, ničeg, NIKOGPrs(4487; 85% of non-emptyPronType): se, ih, mi, ga, im, mu, on, njih, nam, vamTot(13; 0% of non-emptyPronType): svatko, svačemu, svašta
PronType seems to be lexical feature of PRON. 100% lemmas (22) occur only with one value of PronType.
ADV
1463 ADV tokens (17% of all ADV tokens) have a non-empty value of PronType.
The most frequent other feature values with which ADV and PronType co-occurred: Degree=Pos (1463; 100%).
ADV tokens may have the following values of PronType:
Dem(617; 42% of non-emptyPronType): sada, tako, tada, onda, tu, zato, ovdje, stoga, tamo, tolikoInd(56; 4% of non-emptyPronType): ponekad, ikad, nekada, nekoliko, negdje, ikada, nekad, nekako, odnekudInt,Rel(570; 39% of non-emptyPronType): kada, kad, gdje, kako, koliko, zašto, odakleNeg(69; 5% of non-emptyPronType): nikad, nikada, nikako, nigdjeTot(151; 10% of non-emptyPronType): uvijek, svakako, svuda, svugdje, posvudaEMPTY(6976): samo, još, više, također, posto, već, međutim, oko, vrlo, danas
PronType seems to be lexical feature of ADV. 100% lemmas (41) occur only with one value of PronType.
Relations with Agreement in PronType
The 10 most frequent relations where parent and child node agree in PronType:
ADV –[conj]–> ADV (10; 56%),
DET –[conj]–> DET (7; 100%),
DET –[nsubj]–> DET (4; 67%),
PRON –[nsubj]–> PRON (3; 75%),
PRON –[conj]–> PRON (2; 67%),
DET –[appos]–> DET (1; 100%),
DET –[conj]–> PRON (1; 100%),
DET –[fixed]–> DET (1; 100%),
DET –[flat]–> DET (1; 100%),
DET –[nmod]–> DET (1; 100%).