Definition and growth of online translation tools
Vilaró-Soler and Garcia-Riaza (2019) have provided a definition of online translation tools as computer programs that use “artificial intelligence and machine learning algorithms" to automatically translate written text from one language to another Similarly, according to Chen and Ong (2019), online translation tools are "web-based platforms" that employ "machine learning algorithms" to enable users to “automatically translate text from one language to another." Kocbek and Kosem (2019) also offer a definition of online translation tools as “computer programs" that facilitate the translation of text from one language to another through "automated processes", without the need for "human translators."
Online translation tools have become an essential part of modern communication and are widely used in various fields, including business, education, and diplomacy.
These tools provide users with quick and convenient translations of text, documents, and even websites in multiple languages The development of online translation tools has a long history, dating back to the early days of computing.
Machine translation (MT) has a long history, as researchers began working on
"computer programs and algorithms for translating human language into machine language", according to Venuti (2016) Initially, MT tools were based on rule- based translation, where translations were generated by applying linguistic rules and algorithms to the ST Rule-based machine translation (RBMT)'s development can be traced back to the 1960s when researchers began exploring the potential of using computers to translate natural languages During the early stages of RBMT, the translation process relied heavily on dictionaries and grammars, and the rules were often hand-coded by linguists.
However, the limitations of rule-based translation quickly became apparent, and researchers began exploring alternative approaches This led to the development of statistical machine translation (SMT) in the 1990s, which uses statistical models to analyze and translate large amounts of data As O'Brien and Balling (2011) explain, SMT systems acquire the ability to translate through the examination of extensive "parallel text data", resulting in the production of translations that are more natural and precise than those generated by "rule-based systems".
Nguyễn Thị Ngọc Ánh — 11190669 — Business English 61A
The emergence of neural machine translation (NMT) in the early 2010s marked a significant breakthrough in online translation tools NMT uses artificial neural networks to analyze and learn from vast amounts of data, allowing for more accurate and natural-sounding translations As Koehn (2017) notes, NMT has demonstrated superior translation quality compared to "both rule-based and statistical machine translation systems", particularly for intricate and subtle texts.
In conclusion, the development of online translation tools has a rich history that spans several decades, and has been driven by advances in computer science,linguistics, and artificial intelligence The continuous progress of these tools,especially with the introduction of neural machine translation, has significantly enhanced their precision and functionality, rendering them a crucial instrument for communication and cooperation in the present interconnected world.
Disadvantages of online translation tools < s<ôô<ô 20 2.3.3 Some popular online translation f0O0ÌS -s- 5= s=<< ôse 21
Bing Microsoft Translator eo G<SĂ 635 E955589655656596 23 2.3.3.3 VIKHÍTGT-Sẽ(fOF c0 << 5 9 9.9 996 098880.08840888408 896 23 2.3.3.4 Nice TTH1SẽÍOF c <5 <5 9 0 mg 0600996 24 2.4 Previous related r€S€arC: o5 5 5S 9 99.90.0608 689565896 25 2.5 SUMIMALY 1 e
Bing Microsoft Translator is a machine translation tool that was developed by Microsoft and launched in 2007 It has become widely used and is among the most recognized translation tools in the world With support for over 70 languages, Bing Microsoft Translator offers a variety of features, including text, web page, and document translation It also supports real-time translation for Skype calls and Microsoft Office integration.
One unique feature of Bing Microsoft Translator is its "Translator for Microsoft Edge" extension This feature allows users to translate web pages without leaving the page they are currently on It is a convenient tool for users who frequently browse foreign language websites and want to translate content quickly without disrupting their browsing experience Another notable feature of Bing Microsoft Translator is its ability to provide real-time translation for both text and speech.
This feature makes it ideal for communication between people who speak different languages Additionally, the tool offers a feature that allows users to translate entire documents, including Word, PowerPoint, and PDF files This makes it a valuable tool for professionals who work with documents in multiple languages.
While Bing Microsoft Translator has several unique features, like all machine translation tools, it has its limitations Its translations may not always be nuanced and culturally sensitive enough for certain applications, such as legal or medical translations Nonetheless, Bing Microsoft Translator remains a valuable tool for everyday use, given its broad language support and convenient features.
Vikitranslator, an online translation tool created by Viki, a streaming platform for Asian TV shows and movies, supports translation of more than 200 languages and dialects, making it one of the most comprehensive translation tools available It has gained popularity among language enthusiasts and translators due to its development driven by the need to provide accurate translations for Viki's global audience.
The community-driven translation model of Vikitranslator is a distinctive feature that sets it apart from other machine translation tools Vikitranslator's translations are not produced solely by a machine algorithm, but rather by a community of volunteers who are passionate about language and culture This unique approach provides several benefits, including greater accuracy and cultural sensitivity in
Nguyễn Thị Ngọc Ánh — 11190669 — Business English 61A translations Since the volunteers are often native speakers or have a deep understanding of the language and culture, they are able to provide more nuanced and context-specific translations that are not always possible with machine translation alone Another benefit of Vikitranslator s community-driven translation model is the ability for users to suggest and vote on translations This feature enables the community to collectively improve the accuracy of translations, which can be especially helpful for languages with complex grammar or idiomatic expressions.
As with other machine translation tools, Vikitranslator also has its limitations Its translations may lack context and clarity, making it less appropriate for professional and academic purposes Nonetheless, Vikitranslators community- driven approach offers a unique and engaging translation experience.
Nice Translator is a web-based machine translation tool that supports translation of over 60 languages Developed by Nice App, it is a relatively new tool that has gained popularity among users for its user-friendly interface and range of features.
One of the most unique features of Nice Translator is its ability to translate entire web pages with just one click, which is a convenient tool for those who frequently access foreign language websites It saves users time and effort, allowing them to quickly translate the entire page without having to manually translate each section.
Additionally, Nice Translator offers a feature called "Speak Out," which allows users to listen to the translation in the TL, which is an excellent feature for those looking to improve their language skills and pronunciation Another notable feature of Nice Translator is its document translation feature Users can upload documents in various formats, including Word, PDF, and Excel, and the tool will automatically translate the content This feature is especially useful for businesses and professionals who work with documents in multiple languages Additionally, the tool offers a feature that enables users to save translations and access them later, making it a convenient tool for frequent use This feature is particularly useful for users who regularly work with specific phrases or sentences.
Despite its useful features, its translations may lack nuance and cultural sensitivity, which makes it less suitable for professional and academic purposes Overall, Nice Translator is an excellent tool for those who require quick translations and are not too concerned about accuracy or cultural sensitivity While it may not be the most
Nguyễn Thị Ngọc Ánh — 11190669 — Business English 61A accurate translation tool available, its user-friendly interface and range of features make it a popular choice among casual users.
This section delves into existing research on the effectiveness of online translation tools, Vietnamese-English translation, and the translation of political terms to pinpoint gaps and lay the groundwork for the current study.
Numerous studies have investigated the efficacy of well-known online translation tools, such as Google Translate and Microsoft Translator, in translating various languages Nini et al (2018) discovered that as sentence complexity grew, the ability of both translation tools to translate Vietnamese into English diminished.
Nevertheless, for simpler sentences, both tools proved to be effective according to their results (p 75) The study also exposed the difficulties both tools faced with idiomatic expressions and specialized terms, which are often used in political settings.
FINDINGS AND ANALYSTIS Go S5 195B 26 3.1 Common errors made by online translation tools in translating
DISCUSSIONS 70G G5 S9 Ọ ọ Họ 00009100 33 3.2 Degree of accuracy of online translation tools when translating Vietnamese-English political terms in Vietnamese newspapers
The distribution of error types in the online translation of Vietnamese-English political terms from Vietnamese newspapers can be explained that online translation tools are more adept at handling grammatical and cultural aspects, as these tend to follow relatively fixed rules, while they struggle with semantic, lexical, and stylistic aspects, which often require deeper contextual understanding and adaptability.
First, online translation tools can provide a basic level of translation for Vietnamese-English political terms in Vietnamese newspapers, but the high occurrence of semantic and lexical errors suggests that these tools have difficulty comprehending the intended meaning of the terms and selecting the correct words in English In other words, the prevalence of semantic and lexical errors could be the inherent complexity as well as the evolving nature of political terms and the context-specific nature of political discourse Political language often involves the use of specialized vocabulary, idiomatic expressions, and cultural references, which can be challenging for machine translation algorithms to process accurately.
Nguyễn Thị Ngọc Ánh — 11190669 — Business English 61A
In addition, political terms may carry different connotations depending on the context in which they are used, further complicating the translation process.
Next, stylistic errors are quite common when online translation tools handle Vietnamese-English political terms in newspapers, following semantic and lexical errors, due to several factors Because Vietnamese and English have distinct language structures, online translation tools may struggle to maintain the stylistic elements, such as tone and register, that are present in the original text, resulting in stylistic errors Besides, political language is often influenced by the cultural context in which it is used, so online translation tools may have difficulty understanding and accounting for cultural nuances, leading to stylistic errors that can make the translated text sound unnatural or awkward Moreover, political terms in newspapers often contain idiomatic expressions or figurative language that convey specific meanings or emotions, while online translation tools may translate them too literally or failing to find appropriate equivalents in the TL, resulting in stylistic errors.
Lastly, grammatical and cultural errors are the least frequent types of mistakes made by online translation tools when translating Vietnamese-English political terms in newspapers Several factors contribute to the reduced occurrence of these errors Progress in natural language processing and machine learning has resulted in improved language models that can better comprehend and replicate the grammatical structures of languages Consequently, online translation tools are more proficient in addressing grammar-related challenges, leading to a decrease in grammatical errors Additionally, political terms and phrases in newspapers are typically standardized, established, and formal, which lowers the chances of cultural errors Since many of these terms are widely used in global contexts, online translation tools may have access to more training data to learn suitable translations, thereby reducing cultural errors Furthermore, online translation tools have significantly advanced in identifying and translating named entities, such as the names of countries, organizations, and individuals As a result, these tools are less likely to make cultural errors concerning the translation of named entities in political terms and phrases.
In summary, the different levels of prevalence for these error types underscore the necessity for continuous advancements in natural language processing technologies and a deeper comprehension of the limitations inherent in online translation tools when dealing with specialized or context-dependent content It is
Nguyễn Thị Ngọc Ánh — 11190669 — Business English 61A crucial to recognize that while these tools have made significant strides in recent years, they still face challenges in accurately capturing the nuances and intricacies of languages, particularly when translating complex political terms and phrases.
3.2 Degree of accuracy of online translation tools when translating Vietnamese-English political terms in Vietnamese newspapers
The accuracy rates of the four online translation tools for translating Vietnamese- English political terms in Vietnamese newspapers were investigated thoroughly.
As shown in Table 3.3, the accuracy rates for these tools range from 81% to 83%, indicating that while they offer a basic level of translation accuracy for Vietnamese-English political terms in Vietnamese newspapers, there is still potential for improvement.
Number Percentage of Translation Tools of errors accuracy
Table 3.3 — Comparison of errors and accuracy rates for translation tools
Despite its overall popularity and extensive use, Google Translate, with a total of 30 errors, is slightly weaker compared to the other tools Exhibiting an accuracy rate of 81%, Google Translate is the least accurate among the four tools The performance of this tool can be linked to its underlying algorithms and the extensive amount of training data it utilizes, which contributes to enhancing its translation capabilities Nevertheless, the relatively lower accuracy rate suggests that Google Translate may still struggle with comprehending the nuances and context of Vietnamese political terms.
Microsoft Bing Translator has the best performance among the four tools, with a total of 27 errors It boasts the highest accuracy rate of 83%, marginally surpassing the other tools This can be ascribed to the ongoing advancements in its natural language processing algorithms and machine learning models, which have allowed
Nguyễn Thị Ngọc Ánh — 11190669 — Business English 61A it to produce more accurate translations The performance of Bing Translator also implies that it might have better access to relevant training data for political terms or be more successful in integrating this data into its models.
Vikitranslator - with a total of 28 errors, its performance is on par with Nice Translator Both of them display similar accuracy rates of 82% The close accuracy rates of these two tools indicate that they might employ similar translation techniques, algorithms, or training data sets It is plausible that their models have been optimized to manage Vietnamese-English political terms in newspapers more effectively, resulting in their competitive performance.
These results demonstrate that the online translation tools possess a certain level of proficiency in translating Vietnamese political terms in Vietnamese newspapers to English However, it is essential to consider the specific error types and challenges faced by each tool to gain a more comprehensive understanding of their accuracy and limitations.
Distribution of Error Types across Translation Tools
16 mSemantic Lexical Stylistic Grammatical TM Cultural
Google Translate Microsoft Bing Vikitranslator Nice Translator
Figure 3.7 — Distribution of error types across translation tools
By analyzing the data as shown in Figure 3.7, the researcher identifies Google Translate and Vikitranslator both have the highest number of semantic errors, with 14 instances each This suggests that these tools may struggle more with comprehending the intended meaning and context of the terms compared to Bing Translator, which has the lowest number of semantic errors with 12 instances Nice Translator has a slightly lower number of semantic errors at 13 instances,
Nguyễn Thị Ngọc Ánh — 11190669 — Business English 61A indicating that its performance is somewhere between Google Translate/Vikitranslator and Bing Translator in terms of semantic accuracy.
In terms of lexical errors, all tools exhibit a similar performance, with 13 instances for Google Translate, Bing Translator, and Vikitranslator, and 12 instances for Nice Translator This indicates that lexical errors are a common challenge faced by all the tools, possibly due to the presence of homonyms or polysemous words in the Vietnamese language For stylistic errors, all tools have similar numbers, ranging from 9 to 10 instances This suggests that the tools may have difficulty in preserving the original style or tone of the text, resulting in translations that may not be as coherent or smooth as the original text.
Grammatical errors are the least frequent, with only 3 instances for each tool This indicates that the tools have become more proficient in recognizing and reproducing the grammatical structure of languages Lastly, cultural errors are also infrequent, with only | instance for each tool, suggesting that these tools are more adept at handling the translation of named entities and other culturally specific terms.
RECOMMENDA TIONS Go 0009658 39 4.1 Improving the use of online translation (0oÌS <- sô 39 4.2 Combining human expertise with machine translation
Suggesting for the development of translation tooÌs 42 CHAPTER 5: CONCLUUSIONN o0 G00 cọ nọ in 000900 44 REFERENCES 0G G0 Họ cọ cọ ọ 0 TT 00 0.000 009080906 APPENDIX H 0 (G5 sọ cọ TH THỌ 000.000.0000 008000 APPENDIX 2 d G9 9.9 09 0.1809.9.000.08.04004.0809004.009089980909040896
The translation of Vietnamese political terminology into English poses significant challenges due to differences in language structure, cultural references, and unique language features To address these challenges, developers of translation tools need to focus on refining their algorithms and training data sets to better handle semantic, lexical, stylistic, grammatical, and cultural errors This will help improve the accuracy and efficacy of Vietnamese-English political term translations in Vietnamese newspapers.
One of the key approaches to improving the performance of translation tools is to incorporate more specialized data related to political terms and contexts This can be achieved by creating targeted training sets focusing on the unique language features and cultural references specific to Vietnamese politics Developers can also explore using specialized dictionaries or glossaries of political terms in Vietnamese to enhance the accuracy of translations By including more specialized data and resources, these tools can better understand the nuances of political terms and convey intended meanings more accurately.
In addition to refining the algorithms and training data sets, developers can also prioritize the use of machine learning techniques to improve the efficiency of translation tools Machine learning allows tools to continuously learn and adapt to new language patterns and contexts, which is essential when translating political terms with a high level of nuance and complexity By analyzing large volumes of text and recognizing language usage patterns, tools can better comprehend the subtleties of language and improve their translation capabilities For example, Google's translation tool uses machine learning algorithms to improve its accuracy and efficiency, making it a powerful tool for users looking to translate Vietnamese political terms to English.
Nguyễn Thị Ngọc Ánh — 11190669 — Business English 61A
Incorporating human input into the translation process can also improve the accuracy and clarity of Vietnamese-English political term translations One popular approach is to use machine translation as a starting point and have human translators review and edit the output to ensure accuracy This approach, known as
"human-in-the-loop" translation, combines the efficiency of machine translation with the accuracy and context-awareness of human translation This hybrid approach allows for high-quality translations that cater to the requirements of users in the realm of Vietnamese politics.
In summary, to improve the effectiveness of translation tools in translating Vietnamese-English political terms, developers must use a variety of techniques.
These include refining algorithms and training data sets, integrating machine learning methods, and involving human input in the translation process By utilizing these strategies, developers can enhance the precision and efficiency of these tools, meeting the needs of users in the Vietnamese political sphere.
Nguyễn Thị Ngọc Ánh — 11190669 — Business English 61A
This graduation thesis delves into the effectiveness of online translation tools in translating Vietnamese-English political terminology found in Vietnamese newspapers The study is driven by three key research questions: identifying prevalent errors made by online translation tools, determining the accuracy of these tools, and providing recommendations for employing these tools to translate Vietnamese-English political terminology in Vietnamese newspapers To address these questions, the study concentrates on translating political terms from Vietnamese to English, utilizing a sample of 30 articles from three Vietnamese newspapers with a focus on politics during the first quarter of 2023 The research examines four online translation tools: Google Translate, Bing Microsoft Translator, Vikitranslator, and Nice Translator.
The study's methodology involves using the four online translation tools to translate the selected articles and comparing the translations with the newspapers’ published standard translations The errors are categorized and analyzed through tables, charts, and graphs to reveal any trends or patterns The tools' accuracy and effectiveness are evaluated based on the frequency and type of errors and their comparative performance To support the development of research questions and findings, researcher represents the theoretical framework encompassing translation theory, political terminology, and online translation tools The researcher also examines previous related research in the field.
The findings of the study present a comprehensive list of 157 Vietnamese political terms along with their accurate English translations The distribution of error types (lexical, semantic, stylistic, grammatical, and cultural) among the four translation tools is scrutinized, revealing that lexical and semantic errors are the most prevalent, followed by stylistic errors, while grammatical and cultural errors are less common In addition, the accuracy rates of the four translation tools range between 81% and 83%, suggesting a need for improvement Microsoft Bing Translator outperforms the other tools with an accuracy rate of 83%, followed by
Vikitranslator and Nice Translator at 82%, and Google Translate at 81%.
Drawing from the findings, the study offers recommendations for enhancing the utilization of online translation tools, integrating human expertise with machine
Nguyễn Thị Ngọc Ánh — 11190669 — Business English 61A translation, and identifying areas for the development of translation tools This research contributes to understanding the efficiency and limitations of online translation tools in translating Vietnamese-English political terms in Vietnamese newspapers, with potential implications for their application in the rapidly evolving field of machine translation.
A key limitation of this study is rooted in its sample size and selection The research is based on a relatively small sample of 30 articles from three Vietnamese newspapers, which specialize in politics and were published during the first quarter of 2023 Although this sample offers insights into the contemporary political scenario in Vietnam, it may not be comprehensive enough to capture the full spectrum of political terminology used in other sources, such as books, academic journals, or online platforms Moreover, the chosen time frame may not fully represent the variability of political terms in different historical or political contexts As a result, the generalizability of the findings to broader contexts or other sources may be limited.
Moreover, the study's emphasis on translating Vietnamese political terms into English restricts the scope of the investigation, as it is limited to this particular language pair This constraint may hinder the applicability of the findings to other language combinations or multilingual situations, which might present unique challenges and complexities in translation Additionally, the research focuses exclusively on evaluating four online translation tools, possibly not taking into account the performance of other existing or emerging translation technologies that could provide varying degrees of accuracy and efficiency.
To achieve a more thorough understanding of online translation tools' effectiveness, it is recommended that future research not only increase the sample size but also widen the scope beyond the current study's focus on translating Vietnamese political terms into English with a limited sample size To accomplish this, upcoming research could include political documents from various sources, such as books, official government agency websites, academic journals, and other relevant materials This would allow for a deeper assessment of the online
Nguyễn Thị Ngọc Ánh — 11190669 — Business English 61A translation tools' performance in handling political terminology across different contexts and formats.
Additionally, future studies could examine the efficiency of online translation tools in translating political terms across multiple language pairs, extending the research scope beyond Vietnamese-English translations By investigating different domains, subject areas, and genres, subsequent research can offer a more comprehensive understanding of the strengths and limitations of online translation tools, ultimately contributing to the advancement of more precise and effective translation technologies for diverse language pairs and specialized fields.
Nguyễn Thị Ngọc Ánh — 11190669 — Business English 61A
1 Al-Shawi, M S., & Al-Saffar, A A (2013) Errors in English-Arabic translation: Causes and implications Journal of the College of Arts, University of Basrah, 65, 85-108.
2 Baker, M (2011) In other words: A coursebook on translation (2nd ed.).
3 Baker, M (2018) In other words: A coursebook on translation Routledge.
5 Catford, J.C (1965) A linguistic theory of translation: An essay in applied linguistics Oxford University Press.
6 Chen, Y., & Ong, W (2019) The development of online translation tools:
A review Journal of Translation Studies, 3(1), 22-36.
7 Gambier, Y., & van Doorslaer, L (Eds.) (2016) Border crossings:
Translation studies and other disciplines John Benjamins Publishing Company.
8 Gentzler, E (2001) Contemporary translation theories Multilingual
9 Gottlieb, H (2016) Terminology matters! Terminology in theory, practice and research In G Anderman & K Rogers (Eds.), Incorporating corpora:
The linguist and the translator (pp 12-28) Multilingual Matters.
10 Hill, M., & Hughes, J (1998) Cyberpolitics: Citizen activism in the age of the internet Rowman & Littlefield.
11 Hixson, J., & Roberts, R (2014) The language of politics Routledge.
12.Holmes, J S (1972) The name and nature of translation studies In J S.
Holmes (Ed.), Translated! Papers on literary translation and translation studies (pp 67-80) Rodopi.
13 Kim, H., & Poole, K T (2005) A dictionary of the language of politics.
14 Kocbek, S., & Kosem, I (2019) Online translation tools: A review.
International Journal of Translation Studies, 5(1), 54-68.
15.Lefevere, André (1992) Translation, Rewriting, and the Manipulation of
Nguyễn Thị Ngọc Ánh — 11190669 — Business English 61A
16.López-Osormo, M A., García-Sánchez, F., & Rodríguez-González, A.
(2019) Online translation tools and language teaching: An analysis of quality and didactic applicability Education and Information Technologies, 24(3), 2807-2822.
17.Marácz, Laszlo (2013) Machine translation: Challenges in the translation of political texts In M Rossette & R Navarro (Eds.), The Language of Politics: Interdisciplinary Perspectives (pp 38-54) Cambridge Scholars
18 Newmark, Peter (1988) A Textbook of Translation Prentice Hall.
19 Nida, Eugene A., & Taber, Charles R (1969) The Theory and Practice of
20 Nida, Eugene A (1964) Toward a Science of Translating E.J Brill.
21.Nida, Eugene A., & Taber, Charles R (1982) The Bible Translator.
22 Nini, Alessandro, Corradini, Elena, & Guerini, Marco (2018) Evaluating the impact of text complexity on the quality of machine translations In Proceedings of the 11th International Conference on Language Resources and Evaluation (LREC 2018) (pp 74-79) European Language Resources Association (ELRA).
23.Nordquist, Richard (n.d.) Semantic errors in language learning.
ThoughtCo Retrieved from_https://www.thoughtco.com/semantic-errors- definition-1692001
24.O'Brien, Sharon, & Balling, Laura Welcher (2011) Machine translation and global research: Towards improved machine translation literacy in the scholarly community Ethics in Science and Environmental Politics, L1(1),
25.Pym, Anthony (2014) Exploring Translation Theories (2nd ed.).
26.Rojo, Ana Maria, & Ibarretxe-Antuủano, Iraide (Eds.) (2013) Cognitive
Linguistics and Translation: Advances in Some Theoretical Models and Applications De Gruyter Mouton.
27.Roskin, Michael G., & Cord, Robert L (2018) Political Science: An
Nguyễn Thị Ngọc Ánh — 11190669 — Business English 61A
28.Sanchez-Gij6n, Pilar (2007) Legal translation In P Kuhiwczak & K.
Litau (Eds.), A Companion to Translation Studies (pp 146-159).
29.Savigny, Heather, & Marsden, Lee (2017) Doing Political Science and
International Relations: Theories in Action Palgrave Macmillan.
30 Savory, Theodore (1967) The Art of Translation Cape.
31.Singh, Harjinder (2010) Translation: An Advanced Resource Book.
32 Shuttleworth, Mark, & Cowie, Moira (2014) Dictionary of Translation
33 Snell-Hornby, Mary (1988) Translation Studies: An Integrated Approach.
34 Venuti, Lawrence (1995) The Translator's Invisibility: A History of
35.Xu, Yuan (2018) Evaluating online translation tools for translating
Chinese-English legal terms International Journal of Law, Language &
1 Le Thi Huong (2020) The challenges of translating political texts from
Vietnamese into English Journal of Language and Communication, 7(2), 126-137.
2 Nguyen Trung Tinh, Nguyen Thanh Lam, Nguyen Thi Huong (2020).