INTERNATIONAL JOURNAL OF ARTIFICIAL INTELLIGENCE
ISSN: 2692-5206, Impact Factor: 12,23
American Academic publishers, volume 05, issue 04, 2025
https://www.academicpublishers.org/journals/index.php/ijai
page 1734
CREATING A TURKIC LANGUAGES PARALLEL CORPUS FOR THE UZBEK
LANGUAGE CORPUS
Bekchanova Zebo Baxramovna
1st-year Master's student in Computational Linguistics, Urgench State University.
Ko‘palova Shahrizoda Otabek qizi
1st-year Master's student in Computational Linguistics, Urgench State University.
Annotation:
This article highlights the main stages and challenges of creating a parallel
corpus. It also provides an in-depth analysis of the lexical, syntactic, semantic, and pragmatic
features of the language using corpora. Important issues such as taking into account the lexical
differences between Uzbek and other Turkic languages, and the creation of a parallel corpus, are
addressed.
Keywords:
language, language family, Turkic languages, lexical difference, parallel
corpus, lexical, syntactic, semantic, and pragmatic features of the language.
Introduction.
Corpus linguistics is one of the most advanced fields of modern linguistics, playing an
essential role in the scientific study and analysis of language. In recent years, corpus linguistics
has been rapidly developing. The use of parallel corpora not only broadens the scope of applied
linguistic research but also fosters new philosophical perspectives for language teaching studies.
Through corpora, the lexical, syntactic, semantic, and pragmatic features of a language can be
deeply analyzed. Such research is often carried out using parallel corpora. Parallel corpora are
collections of texts with the same content in multiple languages, and their analysis helps identify
similarities and differences between languages. Especially, creating a parallel corpus between
Turkic languages contributes significantly not only to improving language teaching and translation
processes but also to the development of machine translation systems. The scientific, practical,
and technological importance of creating a parallel corpus between Uzbek and other Turkic
languages is substantial.
Main Body.
Creating parallel corpora offers linguists and translators opportunities not only to improve
translation quality but also to better understand linguistic relationships and differences. According
to R. Karimov, parallel corpora are a key tool for studying syntactic, semantic, and pragmatic
similarities between languages, and serve as a foundation for tasks such as linguistic and
extralinguistic tagging, developing parallel corpus algorithms, identifying formal and informal
registers of usage in morphology, syntax, and translation studies. They also enhance the reliability
and objectivity of research that would otherwise rely heavily on linguistic intuition, enabling the
creation of new-generation corpus-based dictionaries and grammars. This highlights the main
purpose of creating parallel corpora. Turkic languages are closely related, making the creation of
a parallel corpus between them highly beneficial. The first step in creating a parallel corpus is to
study the grammatical systems of the languages involved. M. Imrad notes: grammatical analysis
is a crucial stage in the creation of a parallel corpus, enabling the comparison of syntactic structures
across languages. Syntactic and morphological similarities among Turkic languages influence the
methodologies used in translation between these languages. The most reliable data for parallel
corpora are human-translated texts. Texts translated by humans increase the accuracy of the corpus
and serve as a high-quality database for machine translation, offering users higher quality
translations compared to existing machine translation tools. Moreover, parallel corpora are
important tools for identifying lexical relationships between languages. Researchers note that with
INTERNATIONAL JOURNAL OF ARTIFICIAL INTELLIGENCE
ISSN: 2692-5206, Impact Factor: 12,23
American Academic publishers, volume 05, issue 04, 2025
https://www.academicpublishers.org/journals/index.php/ijai
page 1735
parallel corpora, it is possible to identify lexical and semantic similarities, helping linguists
improve translation systems. These corpora assist in analyzing the semantic layers of language and
ensuring accurate, appropriate translations. Considering the lexical differences between Uzbek and
other Turkic languages is crucial in creating a parallel corpus. Another theoretical and practical
aspect of creating parallel corpora is their contribution to a deeper understanding of linguistic
elements in the language learning and translation processes. These corpora also make it possible
to analyze pragmatic differences between languages, as each language has unique contextual
features. Bilingual and multilingual corpora have immense value across various fields. Their
importance in comparative analysis is supported by the views of scholars such as Aijmer and
Altenberg. One important aspect of creating a parallel corpus is maintaining text quality. Accurate
and precise translations ensure the effectiveness of the corpus. It is essential that semantic and
syntactic features of the language are correctly represented in translations. Such corpora are not
only useful for scientific analysis but also for practical applications like machine translation
systems.
Corpus Creation Process.
The following main stages can be identified in creating a parallel corpus between Turkic
languages: in the first stage, the grammatical systems and lexical bases of the languages are
analyzed; in the second stage, translated texts are collected and analyzed; in the third stage, texts
are aligned and refined with the help of specialists to ensure accurate translation. According to
researchers, this process allows for the study of different layers of language and improves
translation quality. When collecting texts, literary and scientific materials are especially preferred,
as they reflect accurate and clear forms of language. Studying the differences between literary and
scientific texts enables comprehensive linguistic analysis when creating a parallel corpus.
Conclusion.
Creating a parallel corpus between Turkic languages is of great significance in linguistics,
translation, and computational linguistics. Through these corpora, it is possible to deeply analyze
similarities and differences between languages. Additionally, parallel corpora simplify the
language learning process and enhance the effectiveness of translation systems. Today, corpora
have become essential tools that save time and effort. Corpus-based language education,
dependency-based parsing, FST technology in morphological analysis, national corpus
development methodologies, corpus-based morphological and semantic analyzers, and neural
technologies based on parallel corpora for machine translation, as well as the development of an
educational corpus for the Uzbek language, are all being actively researched. Parallel corpora are
useful not only in linguistics but also in translation studies, bilingual lexicography, and any field
where language comparison is necessary.
References:
1.
Abdurakhmonova N, Tuliyev U. Morphological analysis by finite state
transducer for Uzbek-English machine translation. Foreign Philology: Language.
Literature, Education. 2018(3):68.
2.
Abdurakhmonova N, Urdishev K. Corpus-based teaching Uzbek as a
foreign language. Journal of Foreign Language Teaching and Applied Linguistics
(JFLTAL). 2019;6(1-2019):131-137.
3.
Abduraxmonova, N. Z. "Linguistic support of the program for translating
English texts into Uzbek (on the example of simple sentences): Doctor of Philosophy
(PhD) dissertation abstract." (2018).
INTERNATIONAL JOURNAL OF ARTIFICIAL INTELLIGENCE
ISSN: 2692-5206, Impact Factor: 12,23
American Academic publishers, volume 05, issue 04, 2025
https://www.academicpublishers.org/journals/index.php/ijai
page 1736
4.
Abdurakhmonova N. The bases of automatic morphological analysis for
machine translation. Izvestiya Kyrgyzskogo gosudarstvennogo tekhnicheskogo
universiteta. 2016;2(38):12-17.
5.
John Hutchins. Machine translation and human translation: in competition
or in complementation. International Journal of Translation, 13(1-2):5–20, 2001.
6.
Johansson, S. Seeing through Multilingual Corpora. Amsterdam:
Benjamins. 2007.
7.
Imrad, M. Parallel Korpuslar Yaratish: Nazariy va Amaliy Asoslar.
Tashkent: Ma'naviyat nashriyoti. 2014.
8.
Karimov Rustam. O‘zbek-ingliz parallel korpusini tuzishning lingvistik va
dasturiy masalalari. Dissertation. Bukhara – 2022.
9.
Q.F. Wen, L.F. Wang, M.C. Liang. Spoken and Written English Corpus of
Chinese Learners. Beijing: Foreign Language Teaching and Research Press, 2005. (In
Chinese)
