1 22 Figure 10 : The efficiencies of using MT for Translation in terms of
Popular MT tools among FFL’s students K61 .- ô
The survey to record data about students’ perception towards the machine translation was participated by 72 students The registered result is analyzed as the following chart
3.1.1 The awareness of using MT tools of FFL’s students
English to Vietnamese translation 15 (20,8%) tracau.vn Babelxl Tflat deepl
Figure 3: The awareness of using MT tools of FFL’s students
It can be seen clearly from the chart above that Google Translate is the most common MT tool of FFL’s students Almost all of the students are familiar with Google Translate, with up to 71 respondents out of 72 iTranslate is ranked second among the given tools in the survey, with about 65% of the students The number of people who have knowledge of Microsoft Bing comprises roughly 60%, followed by the figure for U-dictionary There are only 16 people (22% of respondents) recognizing the Yandex application Besides the five MT tools given, there are other MT tools which are close to FFL’s students in the translation process such as English to Vietnamese translator, tracau.vn, etc.
This is not difficult to explain in the era of information and technology these days. Even the generation in the late 1980s were those born in the cradle of IT and the Internet (Enochsson, Ann-Britt et al Caroline Rizza ; 2009 :7), so for the generation as the object of study today, the use of IT is an integral part of their daily life Accordingly, almost all students know the existence of Machine
Duong Thị Vân Anh— 11190128 — Business English 61C
Translation Sujarwo (2020) also indicated that Google Translate is considerably more popular than the others, perhaps due to Google Translate being older in existence.
3.1.2 Students’ frequency of using MT tools for translating texts
Because of the difference in acquaintances, the students’ frequency of using these tools also varies from one tool to another.
HMM Aways Mil Frequently Ml Sometimes MMMM Rarely MMM Never
Google Translate iTranslate Microsoft Bing Yandex U-dictionary Other (specify)
Figure 4: Students’ frequency of using MT tools for translating texts
Overall, most of the FFL’s students have the habit of using MT tools when translating texts More than half of the participants say that they always use Google Translate in the translation process, which is far greater than the figures for the other MT tools given Furthermore, nearly 30 students use Google Translate frequently to support translating texts The number of participants who rarely or never use Google Translate is only 1 or 2 people The tools ranking behind Google Translate are iTranslate and U-dictionary, with approximately 20 participants who answer “sometimes” equally Microsoft Bing and Yandex are the two MT applications that have the least number of users About half of the respondents never use the two kinds of tools Whereas, the proportion of Microsoft Bing and Yandex users only accounts for a minority, at under 5 people.
It is clear that FFL’s students have a strong preference for using Google Translate among the five given tools during the English Vietnamese translation process. According to Christanta (2020), Google Translate has lately become the most used internet translation service This comes as no surprise as Google Translate is a free translation service that provides instant translations in hundreds of languages It is
Duong Thị Vân Anh— 11190128 — Business English 61C capable of translating words, phrases, and full pages between the languages that we support.
3.3.3 The way of using of MT tools of FFL’s students
MMM Phrases oridioms i Sentences #ΠParagraphs MEM The whole text
Google Translate iTranslate Microsoft Bing Yandex U-dictionary
Figure 5: The way of using of MT tools of FFL’s students
From the chart, it is clear that FFL’s students often use MT tools to translate sentences and paragraphs The number of people who utilize MT tools to translate paragraphs is largest among charts, at over 50 students for Google Translate, and more than 20 students for iTranslate and U-dictionary Translating sentences is the second most popular purpose of FFL's students when they use MT tools The number of participants using Google Translate for this purpose is nearly 40 , followed by iTranslate and U-dictionary, at about 20 people equally.
Except for Google Translate, FFL’s students have a low inclination for translating entire essays Nearly 40 respondents say they translate the whole texts by Google Translate, while the figures for the other tools are under 5 people Likewise, the number of people who use these applications to translate phrases and idioms is only under 20 people for Google Translate, and less than 5 people for the other systems These figures show that the main intention of students when using MT tools seems not to translate phrases, idioms and entire texts.
Duong Thị Vân Anh— 11190128 — Business English 61C
The efficiency of these tools in supporting FFL’s students K61’s
3.2.1 The translation results when using MT tools
40 ME Verylow MM Low (Mi Medium MM High ~~ I Very high
Figure 6: The translation results when using MT tools
From the chart above, FFL’s students indicated their satisfaction with machine translation outputs The number of participants rating translation outputs done by
MT tools at medium level in terms of consistency and meaning is highest among charts of the survey (a half of the students) There was almost no difference in the figures for ‘‘high’’ votes among the two categories, at roughly over 20 answers. Only a few of the students believed MT outputs have very low, low or very high quality This figure comes as no surprise given MT's inability to feel emotion It can be said that both machines “do not care” the situation or context while translating ST to TT The researchers Ahangar & Rahnemoon (2019) also stated that machine translations prefer to delete or decrease target content, showing a tendency to use reduction translation techniques and linguistic compression. Machine translation frequently uses word-for-word translation (Harper, 2018). Therefore, the quality of MT with regard to consistency and meaning is generally rated average by people surveyed.
In terms of grammatical translation, the number of FFL’s students who evaluated
MT translation outputs at medium and high level is equal, at approximately 30 people Meanwhile, the number of votes for ““low”” and ‘‘very high’’ quality of grammatical translation accounted for a small number of 9 votes and 5 votes,
Duong Thị Vân Anh— 11190128 — Business English 61C respectively No one thinks the machine translation is grammatically very low quality These figures show that MT are valued more in terms of grammar than meaning and consistency, at medium and high level.
3.2.2 The efficiencies of using MT for Translation
GEE Strongly disagree MEM Disagree ŒŒ Neutaải MEM Agree MEM Strongly agree khiAkb l MT tools can translate Translation of grammar Translation of terminology Translation of idioms done Using MT tools helps me words in the right context done by MT tools is correct done by MT tools is correct by MT tools is correct save time in translating texts.
Figure 7: The efficiencies of using MT for Translation
According to the chart above, the greatest efficiency of using MT tools is helping users save time It is noticeable that saving time is the only category that receives
“*gtrongly agree’’ votes, with about 18 answers Moreover, there are more than 10 students who agree with the benefits of using MT tools No students disagree or strongly disagree with the fact that using MT tools is a time-saving method It can be concluded the fact that tools help save time is undoubted The reason for this is that machine translation systems can produce translations quickly with a high degree of accuracy There are even translation devices that can do near-to-real- time translations because they will generate translations right after you choose your target language and click the translate button In addition, machines can do translations forever Unlike humans, they are not tired, so they can continue to do the same jobs and more without the rest MT delivers consistent translation quality and can even improve over time.
In terms of grammar, nearly 20 people agree with the correction of MT outputs, double the figure for ““neutral”” answers Only 2 people disagree that grammatical
MT outputs are correct, and no one expresses strong agreement or strong
Duong Thị Vân Anh— 11190128 — Business English 61C disagreement with this statement The figures show that grammatical translation by MT tools is appreciated, as the results demonstrate in the previous section.
The efficiency in contextual and terminological translation have several similarities in proportion The number of “disagree” votes takes the majority in these categories at under 20 students in each option More than 5 people strongly disagree that MT applications can do correctly in terms of context and terminology.
In contrast, the ““ agree’’ and ““ strongly agree’’ options are only chosen by 1€ participant and no one, respectively It can be seen that the given MT tools seem to be not effective in translating terminologies as well as words in the right context.
Translation of idioms done by MT tools is the most ineffective compared to the translation of other patterns From the graph, the correction of idiomatic translation receives the most proportion of ‘‘ strongly disagree’’ and ‘‘disagree’’, at approximately 17 and 13 respondents, respectively No people surveyed agree or strongly disagree with the efficiency of idiomatic translation done by MT tools.
It can be concluded that while comparing overall translation performances of the selected MT tools, it is clear that except grammatical translation, respondents tend to disagree with the accuracy of MT outputs in terms of context, terminologies and idioms This shows that MT still has drawbacks in terminological, idiomatic and contextual translation.
3.2.2.1 The efficiencies of using MT for Translation in terms of context
HMB Verylow MM Low Mi Medium MMB High MMMM Very high
Google Translate iTranslate Microsoft Bing Yandex U-dictionary
Duong Thị Vân Anh— 11190128 — Business English 61C
Figure 8: The efficiencies of using MT for Translation in terms of context
It is noticeable that the number of ‘‘medium’’ votes about the quality of contextual translation of Yandex is nearly 20 people, double the figures for ‘‘low’’ and
““high”° Meanwhile, the same tendency of Google Translate and U-dictionary can be observed in the bar charts Google Translate and U-dictionary receives a large number of ‘‘medium’’ answers for the quality of translating words in the right contexts, at about 23 and 16 answers ‘’Low’’ ratings comprise the second majority in these categories, at more than 20 and 15 students, respectively In contrast, iTranslate and Microsoft Bing share some similarities in proportion The two tools receive a large number of ““low”” votes for the performance of contextual translation, at 20 and 15 people, which are higher than the figure for ‘‘medium’’.
Google Translate and Microsoft Bing are the only two tools that receive more than
10 ‘‘high’’ votes Generally, it can be concluded that a great number of students believe that the level of contextual translation’s accuracy of these machines is low or medium.
According to Lingu et al (2021), Google Translate cannot replace translators due to context and cultural comprehension restrictions.Google translate, or other machine translation, is a program that assists translators with their projects or jobs.
In reality, translators must improve the machine's results However, the quality of contextual translation has been improved compared to results in previous studies. Many ‘’medium’’ votes are given by respondents This is because these MT tools are enhanced, based on the latest system- neural machine translation Swasthi& Jayashree (2020) found that neural machine focused on contextual information more than other machine translation Thus, the current models of these tools are rated medium and low instead of very low or low.
Duong Thị Vân Anh— 11190128 — Business English 61C
3.2.2.2 The efficiencies of using MT for Translation in terms of grammar
MM Verylow MM Low Mi Medium MMB High MMM Veryhigh
Google Translate iTranslate Microsoft Bing Yandex U-dictionary
Figure 9: The efficiencies of using MT for Translation in terms of grammar
The figure illustrates the capacity of the five chosen systems to handle grammatical translation About 35 out of 71 respondents using Google Translate say that translation done by this tool is highly correct with regard to grammar There are
25 people who evaluate that the grammatical translation of Google Translate is medium, which is 5 times higher than figures for the other options of ‘‘very low’’,
‘“‘low’’, and ““very high’’ This means that Google Translate is rated high in providing a translation with reference to grammar.
Meanwhile, there is almost no difference in the trend for iTranslate, Microsoft Bing, Yandex and U-dictionary The four tools receive more number of ‘‘medium’’ than ‘‘high’’ votes in grammatical translation The number of people who rated
“‘medium’’ for iTranslate, U-dictionary, Microsoft Bing and Yandex is more than
25, 24, 20 and 15 respectively, double the figure for ““high”” votes.
The four tools tend to be rated medium with reference to translation of grammar.
This comes as no surprise because currently, the five MT tools, especially Google Translate uses Neural Machine Translation (NMT) technology for most language pairs instead of the previous MT systems' rule-based approach As a result, the grammar will be handled properly by Neural Based Machine Translation, the sophisticated MT technique.
Duong Thị Vân Anh— 11190128 — Business English 61C
3.2.2.3 The efficiencies of using MT for Translation in terms of terminologies
MM Verylow MM Low Mi Medium MMB High MMM Very high
Google Translate iTranslate Microsoft Bing Yandex U-dictionary
Figure 10: The efficiencies of using MT for Translation in terms of terminologies
CONCLUSION eccecceesessseeseeeseeseeeeeeaeceeeeaeeeseceseseeeeaeeeaeeeeeeaeees 26 4.1 Summary of the researCH c1 1311131138 1E 111.1 EeErrre 26 4.2 Limitations of the TS€ATCH - 0 c1 901931191 9 11911 ng gnrưệt 26 4.3 Recommendation for further studies - s5 +5 + +s£ss+svessesexs 27
The study has analyzed the application of MT tools of FFL's students and the efficiency of these machines After introducing typical features of five selected
MT tools which are Google Translate, iTranslate, Microsoft Bing, Yandex and U- dictionary, the research illustrated FFL’s students’ tendency to use these tools. Additionally, the study compared the effectiveness of these five software as well as MT in translating English-Vietnamese documents for FFL’s students.
The research indicated that all of the FFL's students are familiar with the technological tools, especially MT tools that assist them in translating texts The five selected MT tools are used with different frequencies Among them, Google Translate is the most common application, and has mainly been used in their translation process by FFL's students FFL’s students also utilized these tools in various ways, but mostly to translate sentences and paragraphs.
MT tools bring a lot of benefits for students with regard to word processing and speed Using MT tools allows students to save considerable time compared to those who translate by themselves This is because MT tools can translate a large amount of information with just one click In terms of quality of translation outputs, the results of the comparative analysis show that except for grammatical translation, the five MT tools still have several weaknesses related to the fields of context, idioms and terminologies, which prevent machine translation systems from providing accurate translations Google Translate is considered to have more advantage than the other systems in terms of grammatical translating It can be concluded that MT is an imperfect machine translator, and only beneficial to a limited area of translation processes.
This research, like every other research, has significant limitations Because of the time constraints, the questionnaire could not have enough participation of all the seniors in the FFL at NEU During the study, the research just conducts the survey
Duong Thị Vân Anh— 11190128 — Business English 61C by questionnaires to collect data due to lack of the necessary resources As a result, the evaluation of the application of MT tools of participants may lack accuracy and objectivity In addition, for the sake of accessibility, the suggested influencing factors for students’ reference is still subjective.
Within the framework of the report, it is difficult to exploit many aspects of machine translation for document translation MT may be studied in the future in connection to other difficulties such as employing different online translation tools for business or gaining a full grasp of more diverse online translation software applications Further studies should delve deeper into the specific problems faced by students in order to propose more thorough and specific solutions Accordingly, for the practice of translation, future research papers need to systematically study and analyze errors based on the translation outputs provided by students This is a more practical and more thorough approach in proposing solutions.
Duong Thị Vân Anh— 11190128 — Business English 61C
Abdi, H (2019) Translating Culture- Specific Items ( CSIs) as a Conundrum for Iranian M A Translation Students : Considering the Level of Study Journal of New Advances in English Language Teaching and Applied Linguistics,1 (1), 88 — 109
Ahangar, A A., & Rahnemoon, S N (2019) The level of explicitation of reference in the translation of medical texts from English into Persian: A case study on basic histology Lingua Juornal, 228, 102704 Retrieved from: https://doi.org/10.1016/j.lingua.2019.06.005
Alhaisoni, E., & Alhaysony, M (2017) An investigation of Saudi EFL university students’ attitudes towards the use of Google translate International Journal of English Language Education, 72-82.
Allen, J (2004) Case study: implementing MT for the translation of pre-sales marketing and post-sales software deployment documentation, 1-6 The 6th Conference of the Association for Machine Translation in the Americas, AMTA
Anderson, M., & Jiang, J (2018) Teens, social media & technology 2018 Pew Research Center, 31(2018), 1673-1689.
Boilet, C (2007) Corpus pour la TA: types, tailles et problémes associés, selon leur usage et le type de systéme, Revue francaise de linguistique appliquée, Vol. XII, p 25-38.
Brkovic, V., Kardum, R.B., & Togonal, M (2020) Dominant Language Competences in the Contemporary Teaching of History as a
Prerequisite for a Successful Teaching Process International Journal of Instruction, 14(1), 427-444 Retrieved from: https://doi.org/10.29333/IJI.2021.14125A
Catford, J C (1965) A Linguistic Theory of Translation Oxford: Oxford University Press.
Duong Thị Vân Anh— 11190128 — Business English 61C
Chakrawarti, R K., Mishra, H., & Bansal, P (2017) Review of machine translation techniques for idea of Hindi to English idiom translation International journal of computational intelligence research, 13(5), 1059-1071.
Chan, Y S., & Ng, T H (2008) MAXSIM: A maximum similarity metric for ma- chine translation evaluation The Association for Computational Linguistics, Columbus, Ohio, USA
Chang, P.C., Galley, M., & Manning, C.D (2008) Optimizing Chinese word segmentation for machine translation performance The Third Workshop on Statistical Machine Translation, Columbus, Ohio, USA.
Chéragui, M A (2012) Theoretical Overview of Machine translation ICWIT, 160-169.
Cook, G (2010) Translation in language teaching: An argument for reassessment Oxford: University Press.
Dahmash B N., (2020) I Can’t Live Without Google Translate: A Close Look at the Use of GoogleTranslate App by Second Language Learners in SaudiArabia. Arab World English Journal, 11(3), 226-240 Retrieved from: https://doi.org/10.24093/awej/voll Ino3.14
Gu, J., Hassan, H M., Devlin, J., & Li, V C (2018) Universal Neural Machine Translation for Extremely Low Resource Languages arXiv (Cornell University). Retrieved from: https://doi.org/10.48550/arxiv.1802.05368
Guerra, M L (2003) Human translation versus machine translation and full post- editing of raw machine translation output Unpublished MA Dissertation, Dublin City University, Dublin, Ireland.
Harper, K E (2018) Machine translation Soviet and East European Linguistics, October, 133 — 142 Retrieved from: _ https://doi.org/10.4324/978 131567848 1-27
Hatim, B., & Munday, J (2004) Translation: An advanced resource book. Psychology Press.
Duong Thị Vân Anh— 11190128 — Business English 61C
Hees, M V., Kozlowska, P., & Tian, N (2015) Web-based automatic translation: the Yandex Translate API Leiden Institute of Advanced Computer Science
(LIACS) Retrieved from: https://staas.home.xs4all.nl/t/swtr/documents/wt2015_yandex_translate.pdf
Hutchins, W J and Somers, H L (1992) An introduction to machine translation.
Academic Press, London Lingua, 228 Retrieved from: https://doi.org/10.1016/j.lingua.2019.06.005
Junczys D (2016.) Is Neural Machine Translation Ready for Deployment ? A Case
Study on 30 Translation Directions Retreived from: https://www.semanticscholar.org/paper/S YSTRAN's-Pure-Neural-Machine-
Translation-Systems-Crego-Kim/30e0b5cb62be2dc2c I4d3ecae7cccceddfeab4c8
Kaji, H (1988) An efficient execution method for rule-based machine translation. Retrieved from: http://www.aclweb.org/anthology/C/C88/C88-2167.pdf
Kohen, P (2010) “Statistical Machine Translation”, Cambridge University Press, New York, 53
Kiibler, N (2007) La traduction automatique: traduction machine? La Tribune internationale des langues vivantes, Paris: Union des professeurs de langues dans les grandes écoles scientifiques
Kridalaksana, H (2013) Kamus Linguistik (edisi keempat) Gramedia Pustaka Utama, 128
Lin, G C., & Chien, P (2009) Machine Translation for Academic Purposes. Proceedings of the International Conference on TESOL and Translation Retrieved from:https://www.semanticscholar.org/paper/Machine-Translation-for-
AcademicPurposes-LinChien/0939a3ed5b0eecc54e8a4c6dd5609badac9ab935
Liu, Q., Hou, H., Lin, S., & Qian, Y (2005) Introduction to China’s HTRDP machine translation evaluation The MT Summit X, Phuket, Thailand.
Mayoral, R (2001) Aspectos epistenológicos đe la traducció Castelló de laPlana: Universitat Jaume I, 45.
Duong Thị Vân Anh— 11190128 — Business English 61C
Munday, J (2001) ““Introducing Translation Studies: Theories and Applications’’. Psychology Press, 4-5.
Newmark (1988) A Textbook of Translation Hertfordshire: Prentice Hall International, 5.
Nida, E.A and Taber, C.R (1969) The Theory and Practice of Translation Brill
Academic Publishers, Leiden Retrieved from: https://www.scirp.org/%28S %2835 1jmbntvnsjtlaadkposzje%29 %29/reference/re ferencespapers.aspx ?referenceid)73132
Okpor, M D (2014) Machine Translation Approaches: Issues and Challenges. International Journal of Computer Science Issues, 11(5), 159-165 Retrieved from: www.IICSLorg
Shih, C L (2007) Mapping out Students’ Translation Process: An MT-specific Comparative Study Journal of Translation Studies, 163-190 Retrieved from: https://www airitilibrary.com/Publication/alDetailedMesh?docid000024-
Stevanovic, IL, & Radiševié, L (2012) Comparative Analysis of Machine Translation Systems International Journal of Computer Applications, 12(2), 5— 8
Sujarwo, S (2020) Students’ Perceptions of Using Machine Translation Tools In the EFL Classroom Al-Lisan, 6 (2), 230 — 241 Retrieved from: https://doi.org/10.30603/al.v6i2.1333
Swasthi, & Jayashree (2020) Machine Translation Using Deep Learning: A Comparison in Proceedings of International Conference on Artificial Intelligence, Smart Grid and Smart City Applications Retrieved from: https://doi.org/10.1007/978-3-030-24051-6_ 85
Tanner, A (2007) Google seeks world of instant translations Novem-ber 21,
2009 Retrieved from: http://www.reuters.com/article/idUSN192188 1520070328
VTV Digital (2014) iTranslate - Ung dung dich bằng hình ảnh nhờ công nghệ AR.
Retrieved from: https://vtv.vn/cong-nghe/ủtranslate-ung-dung-dich-bang-hinh- anh-nho-cong-nghe-ar-2020081916182624.htm
Duong Thị Vân Anh— 11190128 — Business English 61C
Vuorinen, E (1995) “News translation as gatekeeping”, Translation as Intercultural Communication, Amsterdam-Philadelphia, John Benjamins, 161-172.
Wiratno, T (2003) Analisis Bahasa dari Sudut Pandang Linguistic Semantic Functional Surakarta.
Wuet al (2016) Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation Retreived from: https://www.scinapse.io/papers/2525778437
Ying, Y., Theresia, & Febriana, V (2021) The Advantages of Pleco and Google
Translate Applications in Assisting Mandarin Language Learning Journal of
Physics: Conference Series, 1764(1), 8-11 Retrieved from: https://doi.org/10.1088/1742-6596/1764/1/012106
Zheng, H (2015) A Case Study of Machine Translation: Problems and Suggestions.5(2), 92-99 Retrieved from: https://doi.org/10.5539/ijel.v5n2p92
Duong Thị Vân Anh— 11190128 — Business English 61C
Iam a student at National Economics University Currently, I am doing research on the topic ““An investigation into the application of machine translation tools in translating texts from English to Vietnamese among FFL’s students at NEU”’
I hope you can spend a moment to respond to the following survey Your answers will be a crucial contribution to the result of the study It is guaranteed that all your personal information will be kept confidential and used for research purposes only.
Thank you for participating in this survey!
1 How many machine translation tools do you know? ® 0 ® 1-3 ® 3-5 ® >5
2 Which machine translation tools do you know? ( You can choose more than 1)
@ Google Translate iTranslate Microsoft Bing Babelxl
3 Which machine translation tools do you often use for translating texts? (You can choose more than 1)
English to Vietnamese translationOther (specify):
Duong Thị Vân Anh— 11190128 — Business English 61C
4 How often do you use machine translation tools to translate a text ?
(Please choose the appropriate answer for you)
Always | Frequentl | Sometim | Rarely Never y es mm | | | | | —
5 You use these tools to translate
(Please choose the appropriate answer for you)
Phrases or | Sentences | Paragraphs | The idioms whole text
6 Which of these following problems do you often face in translating texts by machine translation tools?
Lengthy expression Lack of vocabularies Lack of Naturalness Inconsistency
Inappropriate tone (Too formal or casual) Incorrect grammar
7 How do you evaluate translation results when using MT tools in terms of ? (Please choose the appropriate answer for you)
Very low | Low Medium | High Very high
(Y em phan nay 1a sự thông nhất của bản dịch) general?
(Please choose the appropriate answer for you)
Duong Thị Vân Anh— 11190128 — Business English 61C
Translation of terminology done by MT tools is correct
Translation of idioms done by MT tools is correct
Using MT tools helps me save time in translating texts.
9 Rate the efficiency of the MT tools you use in terms of translating words in the right context
Very low | Low Medium | High Very high
Duong Thị Vân Anh— 11190128 — Business English 61C
10 Rate the efficiency of the MT tools you use in terms of translating grammar correctly
Very low | Low Medium | High Very high
11 Rate the efficiency of the MT tools you use in terms of translating terminology correctly
Duong Thị Vân Anh— 11190128 — Business English 61C
12 Rate the efficiency of the MT tools you use in terms of translating idioms correctly
Very low | Low Medium | High Very high má TT | | | —