• Publicationsclose
• Research dataclose
• Research softwareclose
• 2014-2023close
• Open Access close
• Article close
• Slovenščina 2.0: Empirične, aplikativne in interdisciplinarne raziskave close

Seminarja o metodah korpusnega in eksperimentalnega jezikoslovja v Beogradu in Zagrebu

Poletna šola korpusnega jezikoslovja v Lancastru

Prispevek opisuje izdelavo najnovejšega splošnega slovarja poljskega jezika: Veliki slovar poljskega jezika Poljske akademije znanosti (Wielki słownik języka polskiego PAN). Vodilni partner projekta je Inštitut za poljski jezik, pri projektu pa sodelujejo tudi jezikoslovci in leksikografi s številnih poljskih akademskih ustanov. Prispevek na kratko predstavlja zgodovino projekta in glavne značilnosti slovarja. Opisana je tudi organizacija dela in naloge, ki jih opravljajo posamezni člani ekipe. Prispevek dalje prikazuje spletno zasnovo slovarja in dele spletnega vmesnika, npr. različne načine iskanj. V zaključku so navedeni načrti za prihodnost, ki vključujejo tako dodajanje novih iztočnic oziroma novih informacij obstoječim iztočnicam kot tudi izboljšavo leksikografskih postopkov z uporabo sodobnih jezikovnotehnoloških orodij.

V prispevku se obravnava besednovrstna kategorizacija s stališča navajanja besednovrstnih oznak v slovarskem opruiročniku za slovenski jezik na splošni ravni, veljavni ne glede na določeni slovarski koncept. Uvodoma je prikazana svojskost besednovrstnega kategoriziranja v slovnični teoriji, čemur sledi ponazoritev vloge besednovrstnega določanja leksikalnih enot znotraj slovarskega pomenskega opisa. Ob primerih za povedkovnik, členek in izdeležniške tvorjenke je prikazan razvoj besednovrstnega kategoriziranja v doslejšnjih slovenskih slovarjih in na osnovi tega so podane smernice za besednovrstno označevanje leksikalnih enot v prihodnjih slovenskih slovarskih priročnikih.

The Proposal for a Dictionary of Contemporary Slovene, published in May 2013, has stirred many debates in both academic circles and in media. The topic central to all the debates was whether a new dictionary of Slovene should follow the tradition established by the Dictionary of Literary Slovene (published in 1970s), which was based on the structuralist theories of the Prague school, or move away from this tradition. All this lead to differing views on what a dictionary tradition is, and on the role of new lexicographic methods. By analyzing the concepts of the Dictionary of Literary Slovene and the Dictionary of New Slovene Lexis (published in 2012), as well as making an overview of scientific articles dealing with the topic of a concept for a new dictionary of Slovene, this paper attempts to establish which elements of lexicographic theory can be viewed as traditional and which represent innovation in Slovene lexicography. Simultaneously, a concept for a new dictionary is considered from three perspectives: the user, the medium, and the use of language technologies, which would facilitate language description and meet the needs of language community. As the author argues, a new dictionary of Slovene will do well to carefully consider the status of literary language in contemporary Slovene, be corpus driven and user oriented (rather than academic), incorporate various lexicographic findings, e.g. use different approaches to defining (depending on their efficiency at different word classes or categories of words), be digital born, i.e. devised with an online medium in mind, offer updates on a regular basis, and utilize various language technologies, such as automatic example extraction, in its design to facilitate dictionary compilation. Only thus will the new dictionary become a state-of-the-art lexicographic product and a worthy successor to the Dictionary of Standard Slovene.

Computer models based on finite-state transducers are well suited for compact representations of pronunciation lexicons that are used both in speech synthesis as well as in speech recognition. In this paper, we present a finite-state super transducer, which is a new type of finite state transducer that allows the representation of a pronunciation lexicon with fewer states and transitions than using a conventional minimized and determinized finite-state transducer. A finite-state super transducer is a deterministic transducer that can, in addition to the words comprised in the pronunciation lexicon, accept some other, out-of-dictionary words as well. The resulting allophone transcription for these words can be erroneous, but we demonstrate that such errors are comparable to the performance of state-of-the-art methods for grapheme-to-phoneme conversion. The procedure for building finite-state super transducers and a validation of their performance is demonstrated on the SI-PRON pronunciation lexicon. In addition, we also analyze several properties of finite-state transducers with respect to their minimum size obtained by their determinization and minimization. We show that for highly inflected languages their minimum size begins to decrease when the number of words in the represented pronunciation dictionary reaches a certain threshold.

The paper presents the analysis of existing bilingual Slovenian-Hungarian dictionaries, which was made as part of the project aiming to design a concept for a new comprehensive Slovenian-Hungarian dictionary. First, a short historical overview of Slovenian-Hungarian lexicography is provided, including first collections of dialect vocabulary, glossaries, and collections and dictionaries of idioms. Then, an overview of Slovenian-Hungarian and Hungarian-Slovenian dictionaries is made, the first one being published in 1961. The paper then focuses on a comparison on three Slovenian-Hungarian dictionaries, which are currently used by majority of users, namely Slovenian-Hungarian part of the dictionary by Elizabeta Bernjak (1995), Slovenian-Hungarian dictionary by Jože Hradil (1996), and Slovenian-Hungarian part of the Hradil%s bidirectional dictionary. The dictionaries are compared in terms of size, headword list, coverage, headword presentation, grammar information, as well as in terms of other elements of dictionary microstructure such as translations and examples. The discussion section includes an analysis of the coverage offered by the dictionaries of the vocabulary compilled by teachers at bilingual schools in Prekmurje. The results indicate that the coverage of various levels of vocabulary, frequent or rare, is rather poor; as dictionaries are medium-sized and outdated, this is to be expected, however as the analysis shows, some basic concepts are also often not covered (e.g. research, death, allergy). The second part of the discussion is dedicated to the presentation of selected examples of good practice in bilingual lexicography, such as Comprehensive English-Slovenian dictionary Oxford-DZS as the first bilingual dictionary in Slovenia to use the corpus-based approach, as well as offer much more contextual information on the headwords. Also presented are English-Spanish online dictionaries by Oxford University Press and Collins, the focus being on the elements that demonstrate the utilization of the advantages of the digital medium. Prispevek izhaja iz projekta snovanja novega velikega slovensko-madžarskega slovarja, pri katerem je bila za namene priprave koncepta potrebna tudi analiza obstoječih dvojezičnih virov za ta jezikovni par. Tako prispevek najprej ponudi krajši zgodovinski pregled slovensko-madžarskega slovaropisja, od prvih zbirk narečnih besed, glosarjev do frazeoloških zbirk in slovarjev. Sledi pregled dvojezičnih slovarjev za jezikovni par slovenščina-madžarščina in madžarščinaslovenščina. Prispevek se nato osredotoči na primerjavo treh najpogosteje uporabljanih slovensko-madžarskih slovarjev, in sicer slovensko-madžarskega dela slovarja Elizabete Bernjak iz leta 1995, Slovensko-madžarskega slovarja Jožeta Hradila iz leta 1996 ter slovensko-madžarskega dela Hradilovega navzkrižnega slovarja iz leta 2012. Slovarji so primerjani z vidika obsega, geslovnika, predstavitve iztočnic, zaglavja ipd. Primerjani so tudi različni elementi slovarske mikrostrukture, od prevedkov do ponazarjalnega gradiva. Opravljena je kratka študija izrazja, ki ga najdemo v e-zbirkah izrazov in ki naj bi odsevalo potrebe izobraževalnih ustanov na dvojezičnem območju. Temu sledi pregled izbranih primerov dobrih praks sodobne dvojezične leksikografije, kot je pri nas npr. Veliki angleško-slovenski slovar Oxford-DZS, v tujini pa dvojezični slovarji založb Oxford in Collins. Pri pregledu predvsem izpostavimo elemente, ki najbolje kažejo sodobne pristope izkoriščanja prednosti digitalnega medija in so posledično najbolj relevantni za snovanje Velikega slovensko-madžarskega slovarja.

The research of usernames in comments of online articles and on Twitter showed that users who comment online news tend to cover their identity more often than Twitter users. We assume that the reason for this is that online news commentators mostly express their opinion, which may differ from the point of view of the online news portal. Due to anonymity, the opinion can be intensified or even radicalized. Therefore, commentators wish to at least partially conceal their identity. On the other hand, Twitter users frequently reveal their identity in their username, namely by using their real name and/or surname since the main purpose of using Twitter is for private communication and entertainment. On Twitter, it is also possible to find individuals who conceal their identity in order to express their opinion; however, they are less frequent. In usernames, users of both CMC types frequently replace their real name and/or surname with words that relate to their personal features (location, religion, characteristics, activity, horoscope and age) or express their interests (celebrities, nature, trademarks, interests, technology and mathematics, events, food). In this way, their identity is indeed obscure to some extent; however, they still express something about themselves. Users largely cover their identity by choosing an innovative username; they may use a general or neutral word or even construct their username in such way to make it difficult to determine its motivation. The analysis of the collected usernames also revealed some other username characteristics typical for online communication in general. In their usernames, users often play with language because they want the username to be innovative and unique in order to represent them online. Modifications of the text are frequent, i.e. switching letters or replacing them with numbers. Omitting diacritical signs (carons) on č, š, ž is also particular as such letters are often substituted by c, s, z or by letter combinations. Numbers may be used only as indexes or they may indicate the user%s age or year of birth or even refer to a number known from popular culture. Username specific are also the ways of marking word boundaries % words are most often separated by using upper- or lowercase and underscores. Usernames are predominantly written in lowercase. Vzdevki oz. uporabniška imena v računalniško posredovani komunikaciji so reprezentacija uporabnikov v spletu, zato predvidevamo, da se ti zelo potrudijo in jih oblikujejo tako, da jih ta kar najbolje predstavljajo. Uporabniška imena so pogosto zelo inovativna in izkazujejo visoko stopnjo igranja z jezikom, na kar na eni strani vpliva dejstvo, da v obravnavanih tipih računalniško posredovane komunikacije ne sme biti uporabnikov z enakim uporabniškim imenom, na drugi strani pa želja posameznika, da bi bilo njegovo uporabniško ime kar se da inovativno in unikatno. Pri oblikovanju uporabniškega imena uporabniki svojo identiteto zakrivajo oz. razkrivajo na različne načine, in sicer glede na tip računalniško posredovane komunikacije, v katerem sodelujejo. Raziskava uporabniških imen v komentarjih novinarskih prispevkov in na Twitterju je pokazala, da uporabniki, ki komentirajo spletne novice, bolj težijo k zakrivanju identitete kot uporabniki Twitterja. Po drugi strani uporabniki Twitterja v uporabniškem imenu pogosteje razkrijejo svojo identiteto, in sicer tako, da v njem uporabijo ime in/ali priimek.

The paper presents the first part of the study within the research project Opportunities and possibilities of preservation and revitalization of the Slovene language among the Slovene minority community in the neighbouring states, which aims to analyse the situation in Benečija. In the era of global processes, European integrational processes, more favourable and sound legal bases (protection laws) and especially because of bilingual schooling, we assume that in the province of Udine there are the premises for linguistic revitalisation and de-assimilation of younger generations, including the intergenerational transmission of language and/or study of Slovene among the generations where this transmission has been interrupted. The paper introduces the first results of the study which aims to analyse the intergenerational transmission of the Slovene language, the reasons why parents have chosen bilingual schools and their expectations about bilingual schooling. A combination of qualitative and quantitative data has been analysed. Available data concerning Slovenes in Italy has been reviewed. Special stress has been put on data regarding the areas of Udine/Videm and Benečija. In the empirical part, qualitative data has been collected with interviews among parents of children attending the bilingual school in the school year 2016/17. On the basis of the answers, it can be presumed that in the previous generation the intergenerational transmission of Slovene language was often interrupted as schooling started, more exactly nursery school, which marked the transition to Italian. Consequently, bilingual schooling represents an important achievement for the Slovene population of the province of Udine. However, bilingual schooling itself cannot guarantee language preservation and revitalisation. There is the need of more support by means of language planning and policies. It has to be highlighted that all these processes are going on in an environment that is increasingly open and favourable for the Slovene language and bilingualism. Članek predstavlja prvi izsek raziskovalnega dela pri ciljnem raziskovalnem projektu Priložnosti in možnosti za ohranjanje oziroma revitalizacijo slovenščine med slovensko manjšinsko skupnostjo v sosednjih državah, namenjenem preučevanju stanja slovenskega jezika v Beneški Sloveniji. V kontekstu globalizacijskih procesov, evropskih integracijskih procesov, ugodnejših in trdnejših zakonskih podlag (t. i. zaščitnega zakona) ter predvsem zaradi prisotnosti dvojezične šole lahko oblikujemo hipotezo, da v Beneški Sloveniji obstajajo ugodne okoliščine za jezikovno revitalizacijo in deasimilacijo mlajših generacij, kar vključuje tudi ponovno učenje slovenščine med generacijami, pri katerih je medgeneracijski prenos jezika zastal. Iz zbranih podatkov je mogoče izluščiti, da je v prejšnji generaciji medgeneracijski prenos jezika pogosto zastal z začetkom šolanja oz. vstopom v vrtec, ki je zaznamoval prehod na italijanski jezik. Tudi zato je dvojezično šolanje za Beneško Slovenijo pomemben dosežek, vendar samo po sebi še ni zadosten pogoj za ohranjanje ali oživitev medgeneracijskega prenosa jezika. Potrebni so še drugi ukrepi jezikovnega načrtovanja in jezikovne politike. Ob tem gre poudariti, da se ti procesi razvijajo v okolju, ki je bolj naklonjeno slovenščini in dvojezičnosti, kot je bilo v preteklosti.