izpis_h1_title_alt

Oblikovanje kanoničnih oblik pri frazeoloških enotah v strojno berljivem Leksikonu večbesednih enot – uporabniški vidik : magistrsko delo
ID Vidic, Zala (Author), ID Smolej, Mojca (Mentor) More about this mentor... This link opens in a new window, ID Gantar, Polona (Comentor)

.pdfPDF - Presentation file, Download (2,39 MB)
MD5: 268272390B8DD0488A4EBA8F9CFBFAC0

Abstract
Osrednji del magistrske naloge predstavlja obravnava izbranih problemov z oblikovanjem kanoničnih oblik pri frazeoloških enotah (FE) v slovenskem jeziku: izpostavili smo probleme zaporedja sestavin in različnih upovedovalnih možnosti, ki vplivajo na zapis FE. Z anketo smo preverili oceno primernosti zapisov FE glede na predhodno oblikovana pravila in glede na tipične realizacije v korpusu. Ugotovili smo, da je abstraktno zaporedje sestavin v kanonični obliki za stavčne, besedilno zaključene in pragmatične FE ter za FE z leksikaliziranim predmetom v dajalniku neprimerno za vključitev v slovar, zato pa bi jih bilo smiselno oblikovati v skladu z zapisom tipičnih korpusnih oblik. Za nestavčne glagolske FE, ki se po korpusnih podatkih pogosteje pojavljajo s sestavinami v drugačnem zaporedju od predpisanega, smo presodili, da jih je ravno tako bolj smiselno predvideti kot izjeme tega pravila in upoštevati zapis iz tipične korpusne oblike. Za FE, kjer je dajalniška sestavina lahko zapolnjena z različnimi leksikalnimi realizacijami, pa položaja predmeta v dajalniku in tožilniku ne igrata pomembne vloge, zato bi kanonične oblike pogojno lahko sledile pravilu o abstraktnem zaporedju sestavin. Za FE, ki se tipično pojavljajo v pretekliku, vendar so v korpusu prisotne tudi s pojavitvami v drugih slovničnih časih, bi bilo kot nevtralnega zastopnika mogoče uporabiti sedanjik, a ne brez natančnejše obravnave izjem, ki so pri govorcih prepoznane kot neprimerne za vključitev v slovar. Ugotovitve nam bodo pomagale pri oblikovanju ustreznih zapisov FE v nadgradnji leksikona kot tudi pri razumevanju strukturiranosti FE v leksikalnem sistemu slovenščine in dejanski jezikovni rab.

Language:Slovenian
Keywords:slovenščina, frazeološka enota, večbesedne enote, strojno berljiv leksikon večbesednih enot, kanonična oblika, pravila za oblikovanje kanoničnih oblik
Work type:Master's thesis/paper
Typology:2.09 - Master's Thesis
Organization:FF - Faculty of Arts
Place of publishing:Ljubljana
Publisher:[Z. Vidic]
Year:2021
Number of pages:89 str.
PID:20.500.12556/RUL-135126 This link opens in a new window
UDC:811.163.6\'373.7
COBISS.SI-ID:102144771 This link opens in a new window
Publication date in RUL:23.02.2022
Views:1168
Downloads:48
Metadata:XML DC-XML DC-RDF
:
Copy citation
Share:Bookmark and Share

Secondary language

Language:English
Title:Forming canonical forms of phraseological in a machine-readable lexicon of multiword expressions – use'r viewpoint
Abstract:
The main part of the master's thesis deals with the selected problems pertaining to the formation of canonical forms of phraseological units (PU) in the Slovenian language. The issues of the sequence of components and different narrative possibilities that affect the notation of PUs were highlighted. A survey was conducted in order to check speakers’ judgement on the suitability of PU entries in relation to previously formed guidelines and typical realizations in the corpus. We ascertained that the abstract sequence of components in canonical forms for sentence, textual and pragmatic PUs and for PUs with a lexicalized object in the accusative case, is unsuitable for inclusion in a dictionary. However, it is better that the sequence of components is formed according to the notation of typical corpus forms. For non-sentence verbal PUs, which according to the corpus data appear more often with components in a different order from the prescribed, we assessed similarly. Non-sentence verbal PUs are understood as exceptions to this rule and are preferably formed as typical corpus forms, as well. In PUs, in which the position of the component in the dative case can be filled with different lexical realizations, the position of the objects in the dative and accusative case do not play an important role. Therefore, their canonical forms can follow the abstract sequence of components. For PUs that typically appear in a past tense but occur in the corpus in other grammatical tenses as well, the present tense could be used as a neutral representative, but not without further research of exceptions, which are recognized by speakers as unsuitable for inclusion in the dictionary. The findings will help us form appropriate PU entries in the lexicon upgrade as well as understand the structure of PU in the lexical system of the Slovenian language and the actual use of language.

Keywords:Slovenian language, phraseological unit, multiword units, machine-readable lexicon of multiword expressions, canonical form, guidelines for forming canonical forms

Similar documents

Similar works from RUL:
Similar works from other Slovenian collections:

Back