Претрага

Упит dcterms_subject_txt:Natural Language Processing

385 items

  • Part of Speech Tagging for Serbian language using Natural Language Toolkit

    Ranka Stanković, Boro Milovanović (2020)
    Dok se razvijaju složeni algoritmi za NLP (obrada prirodnog jezika), osnovni zadaci kao što je označavanje ostaju veoma važni i još uvek izazovni. NLTK (Natural Language Toolkit) je moćna Python biblioteka za razvoj programa zasnovanih na NLP-u. Pokušavamo da iskoristimo ovu biblioteku za kreiranje PoS (vrsta reči) oznake za savremeni srpski jezik. Jedanaest različitih modela je kreirano korišćenjem NLTK API-ja za označavanje. Najbolji modeli se transformišu sa Brill tagerom da bi se poboljšala tačnost. Obučili smo modele na označenom ...
    obrada prirodnog jezika, mašinsko učenje, neuronske mreže
    ... Index Terms—Natural Language Processing; Machine Learning; Neural Network. I. INTRODUCTION In the last couple of years, a big advancement in the field of Natural Language Processing has occurred. There are state-of- the-art language models that perform exceptionally in various language tasks [1-3] ...
    ... Statistical Part-of-Speech Tagger,” Proc. Sixth Applied Natural Language Processing Conference, Seattle, Washington, USA, 2000 [22] E. Brill, “A simple rule-based part of speech tagger”, Proc. Third conference on Applied natural language processing (ANLC '92), Stroudsburg, Pennsylvania, USA, Mar. 1992 ...
    ... available at: www.dr.rgf.bg.ac.rs Abstract—While complex algorithms for NLP (Natural language processing) are being developed, base tasks such as tagging remain very important and still challenging. NLTK (Natural Language Toolkit) is a powerful Python library for developing programs based on NLP ...

  • An Approach to Efficient Processing of Multi-Word Units

    Cvetana Krstev, Ivan Obradović, Ranka Stanković, Duško Vitas (2013)
    Efficient processing of Multi-Word Units in the course of development of morphological MWU dictionaries is not easy to achieve, especially when languages with complex morphological structures are concerned, such as Serbian. Manual development of this type of dictionaries is a tedious and extremely slow process. To alleviate this problem we turned to our multipurpose software tool, dubbed LeXimir, in the production of lemmas for e-dictionaries of multi-word units. In addition to that, we developed a procedure aimed at making ...
    Natural Language Processing, Grammatical Category, Lexical Representation, MWU, multi-word unit
    ... and Their Automatic Processing. Bulag — Bulletin de Linguistique Appliquée et Générale 32, 73–94 (2007) 19. Savary, A., Rabiega-Wisniewska, J., Wolinski, M.: Inflection of Polish Multi-Word Proper Names with Morfeusz and Multiflex. In: Aspects of Natural Language Processing, Lecture Notes in Computer ...
    ... Cvetana Krstev, Ivan Obradović, Ranka Stanković, and Duško Vitas 1 Introduction Morphological electronic dictionaries of Serbian for natural language processing (NLP) are being developed for many years now. Their development follows the methodology and format (known as DELAS/DELAF) presented for ...
    ... use of finite automata in the lexical representation of natural language. In: Electronic dictionaries and automata in computational linguistics, Lecture Notes in Computer Science, vol. 377, pp. 34–50. Springer (1989) 6. Krstev, C.: Processing of Serbian — Automata, Texts and Electronic Dictionaries ...

  • Terminological and lexical resources used to provide open multilingual educational resources

    Biljana Lazić, Danica Seničić, Aleksandra Tomašević, Bojan Zlatić (2016)
    Open educational resources (OER) within BAEKTEL (Blending Academic and Entrepreneurial Knowledge in Technology enhanced learning) network will be available in different languages, mostly in the languages of Western Balkans, Russian and English. University of Belgrade (UB) hosts a central repository based on: BAEKTEL Metadata Portal (BMP), terminological web application for management, browse and search of terminological resources, web services for linguistic support (query expansion, information retrieval, OER indexing, etc.), annotation of selected resources and OER repository on local edX ...
    otvoreni obrazovni resursi, leksički resursi, obrada prirodnih jezika, terminologija
    ... resources, Natural Language Processing, Terminology 1. INTRODUCTION Natural Language Processing (NLP) has a two-faceted approach to education where one involves e-learning and computer-assisted learning and instruction and the other consists of NLP tools for analysis and use of language by machines ...
    ... Greenhow, J. Sonnevend and C. Agur, Ed. Cambridge, MA: MIT Press, 2016, pp. 22. [3] D. Litman, “Natural language processing for enhancing teaching and learning,” in Proc. Natural language processing for enhancing teaching and learning, 2016, pp. 4170–4176. [4] T. M. Cabré Castellví, Terminology: ...
    ... standardisation so more accurate translations are produced. To summarize above mentioned, terminology now constitues a very important field of Natural Language Processing whilethe work that has been done in the field of terminologyhas become to be an indespensible, widespread used resource. The standards ...

  • Parallel Bidirectionally Pretrained Taggers as Feature Generators

    Ranka Stanković, Mihailo Škorić, Branislava Šandrih Todorović (2022)
    In a setting where multiple automatic annotation approaches coexist and advance separately but none completely solve a specific problem, the key might be in their combination and integration. This paper outlines a scalable architecture for Part-of-Speech tagging using multiple standalone annotation systems as feature generators for a stacked classifier. It also explores automatic resource expansion via dataset augmentation and bidirectional training in order to increase the number of taggers and to maximize the impact of the composite system, which ...
    анотација, обрада природног језика, издвајање обележја, композитне структуре, врста речи

  • FrameNet Lexical Database: Presenting a Few Frames Within the Risk Domain

    Aleksandra Marković, Ranka Stanković, Natalija Tomić, Olivera Kitanović (2021)
    U radu se daje kratak prikaz teorije semantike okvira, na kojoj je zasnovana leksička baza Frejmnet. Predstavljena je koncepcija ove mreže, kao i mogućnosti njene primene. Predstavljena je i leksička analiza koja se primenjuje u projektu izrade Frejmneta i ukazano na razlike između analize zasnovane na okviru u odnosu na analizu zasnovanu na reči. Zatim je prikazano nekoliko povezanih okvira koje prizivaju reči iz domena rizika. U radu je predstavljena i platforma NLTК pomoću koje se mogu koristiti ...
    Srpski jezik, semantika okvira, FrameNet, scenario rizika, rudarski korpus, obrada prirodnog jezika
    ... included. KEYWORDS: Serbian language, frame semantics, FrameNet, risk scenario, mining corpus, natural language processing. PAPER SUBMITTED: 15 July 2021 PAPER ACCEPTED: 6 September 2021 Aleksandra Marković aleksan- dra.markovic@isj.sanu.ac.rs Institute for Serbian Language, SASA Belgrade, Serbia ...
    ... Scientific paper 3 NLTK FrameNet Wrappers NLTK (Natural Language Toolkit) is an easy-to-use natural language pro- cessing Python suite that accesses continually increasing number of corpora and lexical resources. NLTK offers different types of text processing, amongst which are: classification, tokenization ...
    ... in lexicography, it is important to list the most frequent collocates of a LU; collocations are crucial not only in language learning, but also in different natural language processing tasks). Using the word sketch and the collocation risk of (ризик од) as a starting point, a detailed view of the co ...

  • Development of Open Educational Resources (OER) for Natural Language Processing

    Cvetana Krstev, Biljana Lazić, Ranka Stanković, Giovanni Schiuma, Miladin Kotorčević (2015)
    In this paper we present the development of an online course at the edX BAEKTEL platform named “Lexical Recognition in the Natural Language Processing (NLP)”. It is based on the course of the same name for PhD studies at the University of Belgrade, Faculty of Philology. There are not many courses in Computational Linguistics (CL) on OER platforms, and there is none in Serbian either for CL or NLP. We have developed this course in order to improve this ...
    E-Learning, Open Educational Resources, Computational Linguistics, Lexical Resources, edX
    ... l protection, geology and natural language processing, the last being in the focus of this paper. Why Study Natural Language Processing (NLP) and Computational Linguistics (CL)? Natural language processing is the technology for dealing with human language, as it appears in everyday spoken ...
    ... LINGUISTICS AND NATURAL LANGUAGE PROCESSING Computational linguistics (CL) is a theoretical discipline between linguistics and computer science concerned with understanding and modelling the written and spoken language from a computational aspect.[3]Natural Language Processing (NLP) develops ...
    ... (OER) for Natural Language Processing Cvetana Krstev, Biljana Lazić, Ranka Stanković, Giovanni Schiuma, Miladin Kotorčević Дигитални репозиторијум Рударско-геолошког факултета Универзитета у Београду [ДР РГФ] Development of Open Educational Resources (OER) for Natural Language Processing | Cvetana ...

  • Extraction of Bilingual Terminology Using Graphs, Dictionaries and GIZA++

    Branislava Šandrih, Ranka Stanković (2020)
    U nauci, industriji i mnogim istraživačkim oblastima, terminologija se brzo razvija. Najčešće, jezik koji je „lingua franca“ za većinu ovih oblasti je engleski. Kao posledica toga, za mnoga polja termini domena su koncipirani na engleskom, a kasnije se prevode na druge jezike. U ovom radu predstavljamo pristup za automatsko izdvajanje dvojezične terminologije za englesko-srpski jezički par koji se oslanja na usaglašeni dvojezični korpus domena, ekstraktor terminologije za ciljni jezik i alat za usklađivanje delova. Ispitujemo performanse metode na domenu ...
    ekstrakcija terminologije, validacija terminologije, GIZA++, grafovi, Unitex, klasifikacija teksta
    ... Improve Machine Translation in a Computer Aided Translation Environment”. Natural Language Engineering Vol. 23, no. 5 (2017): 763–788 Baldwin, Timothy and Su Nam Kim. “Multiword Expressions”. Handbook of Natural Language Processing Vol. 2 (2010): 267–292 Bouamor, Dhouha, Nasredine Semmar and Pierre Z ...
    ... Translations”. Natural Language Engineer- ing Vol. 23, no. 1 (2017): 31–51 Hamon, T. and N. Grabar. “Adaptation of Cross-lingual Transfer Methods for the Building of Medical Terminology in Ukrainian”. In Proceedings of the 17th International Conference on Intelligent Text Processing and Computational ...
    ... “A Hybrid Approach to Compiling Bilingual Dictionaries of Medical Terms from Parallel Corpora”. Statistical Language and Speech Processing Vol. 8791 (2014): 57–69 Krstev, Cvetana. Processing of Serbian. Automata, Texts and Electronic Dictionaries. Faculty of Philology of the University of Belgrade, ...

  • A Mathematical Learning Environment Based on Serbian Language Resources

    Radojičić Marija, Obradović Ivan, Stanković Ranka, Utvić Miloć, Kaplar Sebastijan (2018)
    In recent years, in line with ever growing usage of Information technology, the learning environments are changing. The amount of available learning materials in various forms has increased. These new environments demand comprehensive learning systems, which enable management of the learning corpus with special attention paid to relevant lexical resources. In this paper we present the concept of a Mathematical Learning Environment in Serbian (MLES), which is based on a corpus of mathematical materials and various lexical resources, enabling ...
    mathematical content, text processing, mathematical formulae
    ... 380–409. [16] Stanković, R., Obradović, I., Utvić, M., (2014). Developing Termbases for Expert Terminology under the TBX Standard. Natural Language Processing for Serbian - Resources and Applications, University of Belgrade, Faculty of Mathematics pp. 12-26. ...
    ... as several resources simultaneously [10]. Although the resources and tools have already been successfully used for a number of various language processing related tasks including query expansion, they need further improvement for management, named entity recognition, terminology extraction ...
    ... multilingual digital libraries of e-journals. Proceedings of the Eighth International Conference on Language Resources and Evaluation (LREC), pp. 1710- 1717. [11] Krstev, C., (2008). Processing of Serbian. Automata, Texts and Electronic Dictionaries Search Engine. Faculty of Philology of the ...

  • Transformer-Based Composite Language Models for Text Evaluation and Classification

    Mihailo Škorić, Miloš Utvić, Ranka Stanković (2023)
    Parallel natural language processing systems were previously successfully tested on the tasks of part-of-speech tagging and authorship attribution through mini-language modeling, for which they achieved significantly better results than independent methods in the cases of seven European languages. The aim of this paper is to present the advantages of using composite language models in the processing and evaluation of texts written in arbitrary highly inflective and morphology-rich natural language, particularly Serbian. A perplexity-based dataset, the main asset for the ...
    General Mathematics, Engineering (miscellaneous), Computer Science (miscellaneous)

  • E-Connecting Balkan Languages

    Cvetana Krstev, Ranka Stanković, Duško Vitas, Svetla Koeva (2009)
    In this paper we present a versatile language processing tool that can be successfully used for many Balkan languages. This tool relies for its work on several sophisticated textual and lexical resources that were developed for most of Balkan languages. These resources are based on several de facto standards in natural language processing.
    Query expansion, e-dictionary, wordnet, proper name, aligned text
    ... versatile language processing tool that can be successfully used for many Balkan languages. This tool relies for its work on several sophisticated textual and lexical resources that were developed for most of Balkan languages. These resources are based on several de facto standards in natural language ...
    ... independent both from Serbian, for which they were initially developed, and from English which seems to be in the background of many natural language processing tools. The main presupposition for the usage of these tools for other languages is the existence of textual and lexical resources developed ...
    ... 2.4 Prolex Database The Prolex project was initiated in 1990s with the study of toponyms in French with aim of appropriately processing proper names in natural language applications [16]. This work has been pursued by development of a Serbian version, which finally led to the design and construction ...

  • Bilingual lexical extraction based on word alignment for improving corpus search

    Jelena Andonovski, Branislava Šandrih, Olivera Kitanović (2019)
    Library and Information Sciences,Computer Science Applications

  • An Italian-Serbian Sentence Aligned Parallel Literary Corpus

    Saša Moderc, Ranka Stanković, Aleksandra Tomašević, Mihailo Škorić (2023)
    This article presents the construction and relevance of an Italian-Serbian sentence-aligned parallel corpus, delving into the aligned sentences in order to facilitate effective translation between the two languages. The parallel corpus serves as a valuable resource for language experts, researchers, and language enthusiasts, fostering a deeper understanding of linguistic nuances and cultural expressions. By bridging the gap between Serbian and Italian, this corpus opens new avenues for cross-cultural communication and collaboration, and ultimately contributes to the improvement of language-related ...
    Aligned corpus, parallel corpus, Serbian, Italian, literature

  • Managing mining project documentation using human language technology

    Aleksandra Tomašević, Ranka Stanković, Miloš Utvić, Ivan Obradović, Božo Kolonja (2018)
    Purpose: This paper aims to develop a system, which would enable efficient management and exploitation of documentation in electronic form, related to mining projects, with information retrieval and information extraction (IE) features, using various language resources and natural language processing. Design/methodology/approach: The system is designed to integrate textual, lexical, semantic and terminological resources, enabling advanced document search and extraction of information. These resources are integrated with a set of Web services and applications, for different user profiles and use-cases. Findings: The ...
    Digital libraries, Information retrieval, Data mining, Human language technologies, Project documentation

  • Digital Library From A Domain Of Criminalistics As A Foundation For A Forensic Text Analysis

    Dalibor Vorkapić, Aleksandra Tomašević, Miljana Mladenović, Ranka Stanković, Nikola Vulović (2017)
    U ovom radu predstavljen je model koji omogućava prikupljanje, pripremu, opis metapodataka, upravljanje i eksploataciju, uključujući pretragu punog teksta dokumenata iz domena kriminalistike napisanih na srpskom jeziku. Predloženi pristup primenjuje se na veb portalu koji sakuplja različite tekstove nastale iz časopisa Akademije za kriminalistiku i policijske studije, Krivičnog zakona Srbije, konferencija „Tara“ i „Reiss“, kao i iz nekih doktorskih disertacija vezanih za ovu oblast istraživanje. Nakon obrade teksta, korpus koji sadrži preko 5500 stranica običnog teksta, kreiran je i ...
    Omeka, Wordnet, pretraga punog teksta, morfološka i semantička pretraga teksta, proširenje upita
    ... LINGUISTICS The linguistic study of forensic texts is a part of the field of Natural Language Processing, which includes text types classification and syntax and semantic analysis of texts written in a natural language. Various texts are subject of the study: Acts of Parliament (or other law-making ...
    ... Krstev, I. Obradović & D. Vitas Natural Language Processing for Serbian – Resources and Application, 1-11. Matematički fakultet, Beograd. 21 Mladenović, M., Mitrović, J., Krstev, C., & Vitas, D. (2015). Hybrid Sentiment Analysis Framework For A Morphologically Rich Language. Journal of Intelligent Information ...
    ... not in Serbian language was removed, as well as tables, figures, references and links, as usual preparation for corpus processing. After this preparation, the text collection contained 5,500 pages of plain text, in A4 format, which was used for further text analysis and processing. For digital objects ...

  • Towards Automatic Definition Extraction for Serbian

    Ranka Stanković, Cvetana Krstev, Rada Stijović, Mirjana Gočanin, Mihailo Škorić (2021)
    U radu su prikazani preliminarni rezultati automatske ekstrakcije kandidata za definicije rečnika iz nestrukturiranih tekstova na srpskom jeziku u cilju ubrzanja razvoja rečnika. Definicije u rečniku Srpske akademije nauka i umetnosti (SANU) korišćene su za modelovanje različitih tipova definicija (opisnih, gramatičkih, referentnih i sinonimskih) koje imaju različite sintaksičke i leksičke karakteristike. Korpus istraživanja sastoji se od 61.213 definicija imenica, koje su analizirane korišćenjem morfoloških e-rečnika i lokalnih gramatika implementiranih kao pretvarači konačnih stanja u paketu za obradu korpusa otvorenog ...
    ... Conference on Empirical Methods in Natural Language Processing, pp. 780-790. Tissier, J., Gravier, C., & Habrard, A. (2017). Dict2vec: Learning Word Embeddings using Lexical Dictionaries. In Proceeding of the Conference on Empirical Methods in Natural Language Processing (EMNLP 2017), Sep 2017, Copenhague ...
    ... definitions into consistent word embeddings. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pp. 1522-1532. Barnbrook, G. (2002). Defining Language, A local grammar of definition sentences, Studies in Corpus Linguistics, (Vol. 11). John Benjamins Publishing ...
    ... In: 1st Workshop on Recent Advances in Slavonic Natural Language Processing, 2007, pp. 65–70. SASA Dictionary: Речник српскохрватског књижевног и народног језика САНУ, I–XXI [The Dictionary of the Serbo-Croatian Standard and Vernacular Language] (1959–2020). Београд: Институт за српски језик САНУ ...

  • A Twitter Corpus and Lexicon for Abusive Speech Detection in Serbian

    Danka Jokić, Ranka Stanković, Cvetana Krstev, Branislava Šandrih (2021)
    Uvredljivi govor na društvenim medijima, uključujući psovke, pogrdni govor i govor mržnje, dostigao je nivo pandemije. Sistem koji bi bio u stanju da detektuje takve tekstove mogao bi da pomogne da internet i društveni mediji postanu bolji virtuelni prostor sa više poštovanja. Istraživanja i komercijalna primena u ovoj oblasti do sada su bili fokusirani uglavnom na engleski jezik. Ovaj rad predstavlja rad na izgradnji AbCoSER-a, prvog korpusa uvredljivog govora na srpskom jeziku. Korpus se sastoji od 6.436 ručno označenih ...
    uvredljivi jezik, govor mržnje, srpski, tviter, leksikon, korpus
    ... arXiv:1709.10159. 38 Anna Schmidt and Michael Wiegand. A survey on hate speech detection using natural language processing. In Proceedings of the Ąfth international workshop on natural language processing for social media, pages 1–10, 2017. 39 Alessandro Seganti, Helena Sobol, Iryna Orlova, Hannam Kim ...
    ... with abusive triggers extracted from the AbCoSER dataset. 2012 ACM Subject ClassiĄcation Computing methodologies → Natural language processing Keywords and phrases abusive language, hate speech, Serbian, Twitter, lexicon, corpus Digital Object IdentiĄer 10.4230/OASIcs.LDK.2021.13 Funding Linked ...
    ... on a Common Natural Language Processing Paradigm for Balkan Languages, pages 15–22, 2007. LDK 2021 https://www.aclweb.org/anthology/2020.lrec-1.401.pdf https://www.aclweb.org/anthology/2020.globalex-1.1.pdf https://www.aclweb.org/anthology/2020.globalex-1.1.pdf 13:16 Building Language Resources for ...

  • Two approaches to compilation of bilingual multi-word terminology lists from lexical resources

    Branislava Šandrih, Cvetana Krstev, Ranka Stanković (2020)
    In this paper, we present two approaches and the implemented system for bilingual terminology extraction that rely on an aligned bilingual domain corpus, a terminology extractor for a target language, and a tool for chunk alignment. The two approaches differ in the way terminology for the source language is obtained: the first relies on an existing domain terminology lexicon, while the second one uses a term extraction tool. For both approaches, four experiments were performed with two parameters being ...
    Linguistics and Language,Software,Artificial Intelligence,Language and Linguistics

  • The Nooj System as Module within an Integrated Language Processing Environment

    Ranka Stanković, Duško Vitas, Cvetana Krstev (2008)
    NooJ, electronic dictionary, lexical resources
    ... System as Module within an Integrated Language Processing Environment Ranka Stanković, Duško Vitas, Cvetana Krstev Дигитални репозиторијум Рударско-геолошког факултета Универзитета у Београду [ДР РГФ] The Nooj System as Module within an Integrated Language Processing Environment | Ranka Stanković, Duško ...
    ... as the employees' publications. - The Repository is available at: www.dr.rgf.bg.ac.rs The NooJ system as module within an integrated language processing environment Ranka Stanković, ranka@rgf.bg.ac.yu Duško Vitas, vitas@matf.bg.ac.yu Cvetana Krstev, cvetena@matf.bg.ac.yu 1. Introduction ...
    ... http://www.lisa.org/tmx/ Vitas D., Krstev C., Obradović I., Popović Lj., Pavlović-Lažetić G.: Processing Serbian Written Texts: An Overview of Resources and Basic Tools., Workshop on Balkan Language Resources and Tools, Thessaloniki, Greece, eds, S. Piperidis and V. Karkaletsis, pp. 97-104, ...

  • Knowledge and Rule-Based Diacritic Restoration in Serbian

    Cvetana Krstev, Ranka Stanković, Duško Vitas (2018)
    In this paper we present a procedure for the restoration of diacritics in Serbian texts written using the degraded Latin alphabet. The procedure relies on the comprehensive lexical resources for Serbian: the morphological electronic dictionaries, the Corpus of Contemporary Serbian and local grammars. Dictionaries are used to identify possible candidates for the restoration, while the dataobtainedfromSrpKorandlocalgrammarsassistsinmakingadecisionbetween several candidates in cases of ambiguity. The evaluation results reveal that,dependingonthetext,accuracyrangesfrom95.03%to99.36%,whilethe precision (average 98.93%) is always higher than the recall (average 94.94%).
    diacritic restoration, morphological dictionary, corpus, word n-grams, local grammars
    ... resources having the format of the RuThes thesaurus (Loukachevitch and Dobrov, 2014) and created for automatic processing of documents in information- analytical systems and natural language processing. These resources are linguistic ontologies uniting some principles of their organization from WordNet, ...
    ... Knowledge Engineering Review, 23(1):101–115. Wilks, Y. (2009). Ontotherapy, or how to stop worrying about what there is. Recent advances in natural language processing V, pages 1–20. Will, L. (2012). The iso 25964 data model for the structure of an information retrieval thesaurus. Bulletin of the Association ...
    ... presented the RuThes family of Russian thesauri intended for natural language process- ing and information retrieval applications. RuThes-like thesauri include, besides RuThes, Sociopolitical thesaurus, Security Thesaurus, and Ontology on Natural Sciences and Technology. The RuThes format is based on three ...

  • Advancing Sentiment Analysis in Serbian Literature: A Zero and Few-Shot Learning Approach Using the Mistral Model

    Milica Ikonić Nešić, Saša Petalinkar, Mihailo Škorić, Ranka Stanković, Biljana Rujević (2024)
    Ova studija predstavlja analizu sentimenta srpskih starih romana iz perioda 1840-1920, koristeći veliki jezički model (LLM) Mistral za tehniku učenja sa zasnovani na takozvanim "zero" i "few-shot" pokušajima. Glavni pristup uvodi inovacije osmišljavanjem istraživačkih upita (promptova) uključuju tekst sa uputstvom za klasifikaciju bez primera i na osnovu nekoliko primera, omogućavajući jezičkom modelu da klasifikuje osećanja u pozitivne, negativne ili objektivne kategorije. Ova metodologija ima za cilj da pojednostavi analizu osećanja ograničavanjem odgovora, čime se povećava preciznost ...
    zero-shot, few-shot, sentiment, Serbian, Mistral model