An evaluation of middle aged hanoi tourists’ experience when using google translate for travelling

VIETNAM NATIONAL UNIVERSITY, HANOI UNIVERSITY OF LANGUAGES AND INTERNATIONAL STUDIES FACULTY OF ENGLISH LANGUAGE TEACHER EDUCATION GRADUATION THESIS AN EVALUATION OF MIDDLE-AGED HA

VIETNAM NATIONAL UNIVERSITY, HANOI

UNIVERSITY OF LANGUAGES AND INTERNATIONAL STUDIES

FACULTY OF ENGLISH LANGUAGE TEACHER EDUCATION

GRADUATION THESIS

AN EVALUATION OF MIDDLE-AGED HANOI TOURISTS’ EXPERIENCE WHEN USING

GOOGLE TRANSLATE FOR TRAVELLING

Supervisor: Nguyễn Thị Diệu Thúy, M.A

Student: Đoàn Huyền Trang

Class: QH2016

HÀ NỘI - 2020

ĐẠI HỌC QUỐC GIA HÀ NỘI

TRƯỜNG ĐẠI HỌC NGOẠI NGỮ KHOA SƯ PHẠM TIẾNG ANH

KHÓA LUẬN TỐT NGHIỆP

ĐÁNH GIÁ TRẢI NGHIỆM SỬ DỤNG GOOGLE TRANSLATE CỦA DU KHÁCH HÀ NỘI TRUNG

NIÊN KHI DU LỊCH

Giáo viên hướng dẫn: Nguyễn Thị Diệu Thúy

Sinh viên: Đoàn Huyền Trang

Khóa: QH2016

HÀ NỘI - 2020

Signature of Approval:

Supervisor’s Comments & Suggestions

ACCEPTANCE

I hereby state that I: Đoàn Huyền Trang, Class QH2016.E22, being a candidate for a degree of Bachelor of Arts (Honour programme of Translation and Interpreting) accept the requirements of the College relating to the retention and use

of Bachelor’s Graduation Paper deposited in the library

In terms of these conditions, I agree that the origin of my paper deposited in the library should be accessible for the purposes of study and research, in accordance with the normal conditions established by the librarian for the care, loan or reproduction of the paper

Signature

Doan Huyen Trang

Date: May 4, 2020

ACKNOWLEDGEMENTS

This part is dedicated to the following people without whom I could not have finished this research paper

First and foremost, I would like to express my deepest and most sincere gratitude

to my supervisor and former teacher, Ms Nguyen Thi Dieu Thuy (M.A.) I am forever grateful for her timely tutelage and support, both academically and emotionally Had it not because of her I could never shape this research paper as it

I also want to deliver my utmost love and gratitude towards my families as well, for always being big supporters

Finally, many thanks to myself for always staying focused on what I have decided

ABSTRACT

The purpose of this research paper is to examine the limitations and effectiveness as well as the user experience by middle-aged participants towards Google Translate while traveling, an issue which is still new and has a lot of rooms

to investigate and address

With the aim of enlightening people about the strong existence of Google Translate, its limitations, and weaknesses among middle-aged participants, a set of interviews and questions was posed to middle-aged adults who are currently living

in Hanoi and have used Google Translate when traveling From the figures and evidence collected from questionnaires and interviews, it was found out the main purposes of using Google Translate when traveling is to have assistance at the hospital, on the road and at the restaurants The study also pointed out that although some Google Translate features have existed for a long time, it was speech translation and image translation to be used the most by middle-aged participants for technical and quality reasons Moreover, this study managed to corner the limitations and effectiveness of Google Translate Specifically, limitations and effectiveness regarding naturalness and accuracy were addressed in this study

From this reality, recommendations are made based on the existing limitations and middle-aged participants to reduce the unpleasant experience and further improve the effectiveness of Google Translate and hopefully other machine translation applications

1.1 Statement of research problem and rationale 3

1.3 Previous studies on efficiencies and limitations of Google Translate 9

4.1 Question 1: What are the main purposes of using Google Translate and what features are mostly used by middle-aged tourists? 14 4.2 Question 2: In terms of content and naturalness, what are the efficiencies and limitations that users find when using Google Translate to translate English to

CHAPTER 6: REFERENCES 34

CHAPTER I: INTRODUCTION

With the advent of the Internet in the 1990s and the commensurate rapid growth of information and communication technology, Google Translate has taken a further step towards providing more informed and reliable products for the clients Google Translate is a popular translation service developed by Google in April

2006 Since 2012, Google has announced that more than 200 million people actively use Google Translate monthly (Kerr, 2012) Google translates multiple forms of texts and media such as words, phrases, and webpages and it offers a wide range of services including translating multiple forms of text and media such as signs, speech, images, and videos Additionally, Google Translate functions include written translation, website translation, document translation, speech translation, mobile app translation, image translation, handwritten translation Google Translate is available for texting on mobile phones, in several internet browsers for any Web page, YouTube video captions, and speech-to-speech "conversation mode" on smartphones

The translation method that Google used to adopt is the Statistical Machine Translation (SMT) and later in November 2016, Google has transitioned its translating method to a system called “Neural Machine Translation”, which marked

a significant advance on previous methods The Neural Machine Translation method adopts a technology called deep learning, a system which utilizes the users’ translation to translate the whole sentences at a time and ensures more adequate accuracy

As tourism is an ever-developing industry, the need to have translation applications for tourists is rapidly increasing Google Translate research scientist Franz Och had mentioned that the majority of Google Translate users access the service while traveling (Kerr, 2012)

In Vietnam, Google Translate is among the most commonly used applications that can help users translate from Vietnamese to several languages and vice versa (TNK, 2017) Google Translate is very convenient since it belongs to the Google ecosystem and it compiles several functions to help travellers With availability for approximately 100 languages and works both online and offline, Google Translate is evolving and expanding its operations in this potential market

2 STATEMENT OF RESEARCH PROBLEM & RATIONALE

2.1 Research rationale

The rationale of this research paper stems from three perspectives Firstly, from the role of users, the researcher is an active user and finds Google Translate quite helpful when travelling However, opinions about this translation tool may vary among other age groups, for example the elder people or middle-aged adults Therefore, this research is expected to resolve the question about such matter Secondly, the researcher wants to investigate the effectiveness of Google Translate when being used by people with different English proficiency Specifically, the researcher would like to conduct study based on the experience of middle-aged Vietnamese living in Hanoi who do not have adequate English proficiency to exchange information in another language

From the perspective of a student studying translation and interpreting, translation applications may pose as a threat since it can make translation job become unnecessary as the society is stepping into the 4.0 Era where digital translation machine inventions are accelerating The researcher wanted to take this opportunity to discover this new translation applications` features and effectiveness Also, it is expected that this study can help discovering the potential of future interpreters and translators and understanding the challenges translators may face

2.2 Research problem

Firstly, despite being commonly used, Google Translate is still criticized by users for its inaccuracy and translation mistakes In some cases, tourists often find using Google Translate difficult because the translated versions cannot convey the meaning of the source texts or it provides inappropriate translation, which sometimes can be disturbing and inconvenient

Secondly, there has not been much research that were based on the perspective of users Previous research about Google Translate only focus on the translation mistakes from the viewpoint of researchers using theoretical framework As Google Translate is a digital service for public use, the researcher believes that an approach from the role of users and the study of users’ experience may provide a different viewpoint towards this issue

Finally, analysis on translating Vietnamese to English are still limited while the differences between English and Vietnamese are quite significant, which may pose many questions surrounding translations Based on that mindset, the researcher believes that there can be rooms for investigation

3.1 Research objectives

First and foremost, this research aims to investigate the efficiencies and limitations of Google Translate as perceived by middle-aged Hanoi tourists when travelling Specifically, the research will analyse the experience of users when using Google to translate certain text types by using different data collection instruments Based on that analysis, the researcher expects to have a generalisation about Google Translate performance in various settings Secondly, the researcher wants to have deeper understanding about the reasons why mistakes happened in such contexts during the language translation process In that way, it is hoped that this research paper can provide suggested solutions for the users so that they can fully optimize Google Translate

3.2 Research questions

The aforementioned objectives can be summarized by the following questions:

1 What are the main purposes of using Google Translate and what

features are used by middle-aged tourists?

2 What are the efficiencies and limitations that users find when using Google Translate for travelling?

4 Significance of the study

With this study, the researcher can provide an insightful understanding about machine translation applications, specifically an in-depth look into two functions of Google Translate Furthermore, the significance of this paper is that it shows the experience of middle-aged Vietnamese living in Hanoi, an interesting and potential group of customers towards Google Translate as a digital products This paper is also expected to make generalisation about Google Translate’s translation quality in certain text types and settings What’s more, this research paper can become a contribution to future research about machine translation software or any other research that shares the same concern

CHAPTER II: DEVELOPMENT CHAPTER 1: LITERATURE REVIEW

In this research, further study about related terms and concepts will be reviewed with the hope to coin out the definitions or concepts that will be considered when conducting this research In general, there are 4 definitions and concepts which requires for previous findings: approaches to machine translation, speech to speech translation, instant camera translation, and previous studies about Google Translate

efficiencies and limitations

1.1 Overview of machine translation

Although machine translation was first developed in the 9th century, the first definition of machine translation only appeared in the 17th century In 1629, René Descartes proposed a universal language, with equivalent ideas in different tongues sharing one symbol Machine translation is the task of automatically converting source text in one language to text in another language In a machine translation task, the input already consists of a sequence of symbols in some language, and the computer program must convert this into a sequence of symbols in another language

(p.98, Deep Learning, 2016)

There are several approaches to machine translation In general, there are 4 approaches relating to machine translation, which are rule-based machine translation (RBMT), statistical machine translation (SMT), example-based machine translation (EBMT) and neural machine translation (NMT) Among these approaches, rule-based machine translation (RBMT) is the first type of machine translation to be adopted To start with, in a study about rule-based machine translation, Lagarda (2009) had coined that rule-based machine translation is a system which focuses on the syntactic and semantic analysis of both the source and target language It collects the information about main semantic, morphological, and syntactic regularities of each language On the other hand, example-based machine translation, Sergei (1989) defined example-based machine translation as a system where example of previous translation would

be kept and applied in similar texts and documents It acts like a corpus storing examples and greatly depends on users’ translations

Later on, statistical machine translation (SMT) appeared and was introduced as

a translation engine with a very large volume bilingual (source texts and their translations) The system searches for statistical correlations between source texts and translations Based on the topic and main focus of this paper, the researcher believes that it is not suitable to consider rule-based and neural machine translation definition because of its drawbacks Firstly, with rule-based machine translation, since it depends on the semantic and morphologic of texts, this approach cannot work well with informal texts and idiomatic expressions such as slang or colloquial terms

(Lagardaa, 2009)

As this paper focuses on the use of Google Translate for travelling, colloquial terms and informal speech may appear a lot so this definition is not a good match to the research Therefore, this research paper only chooses example-based and statistical machine translation as the two approaches for further analysis in the study

1.2 Two functions of Google Translate

As this paper has mentioned, Google Translate operates a wide range of functions, among which are two outstanding functions: speech translation and instant camera translation These two functions are the most widely used for tourists when travelling since it is time-saving and very convenient The use of these translations are also the two focal matters for the researcher to analyse Therefore, it would be a loss

not to include any findings or knowledge about it

1.2.1 Image translation

Instant camera translation is a relatively new features, thus the findings on this concept are very limited In general, instant translator is designed to make it extremely easy to translate words from your papers, letters and other sources (Gu, 2019) It includes both recognition of the word (OCR) pointed by your phone camera in selected source language and translation to a set of desired target languages all instantly Google

Translate is the leading application that announce the use of this feature Google Translated previously claimed that this machine translation application to make instant translation by just pointing your camera lens at the foreign text In May 2014, Google acquired Word Lens to improve the quality of visual and voice translation (Petrovan, Bogdan, 2015) It is able to scan text or picture with one's device and have it translated instantly Moreover, the system automatically identifies foreign languages and translates speech without requiring individuals to tap the mic button whenever speech translation

is needed.(McGuire, Nick, 2018) What enables Google Translate to function is Word Lens, a technology that enables users point your smartphone to a foreign language text and have it instantly replaced with your language of choice, right on the screen Street signs, restaurant menus, product labels, there are tons of situations users could find it useful.This feature should make for much more natural conversations, akin to having a human interpreter, translating the back and forth in real time

1.2.2 Speech translation

Speech to speech translation is one of the functions that Google Translate provides To use this function, users first have to give access to microphone on mobile phones When installing and using Google Translate, users can press the microphone icon on the applications and start making their speech Google Translate will process the input in source text and provide the output as the translation in the target language Speech-to-speech translation systems have been developed over the past several decades with the goal of helping people who speak different languages to communicate with each other Such systems have usually been broken into three separate components: automatic speech recognition to transcribe the source speech as text, machine translation to translate the transcribed text into the target language, and text-to-speech synthesis (TTS) to generate speech in the target language from the translated text (Jia, Weiss, 2019)

1.2.3 Previous studies on efficiencies and limitations of Google Translate

A few studies have attempted to assess Google Translate’s accuracy (e.g., Aiken, et al., 2009b) In the past, previous studies have focused on the use of machine translation for specialized fields such as medical and science (Patil, 2014) Specifically, in Patil’s work on the efficiency and limitation of Google Translate in medical communication, it was pointed out that in medical conversation, about 42%

of medical specialized terms was mistranslated Some studies even attempt to analyze the efficiency of Google Translate by inviting participants who are advanced translators to assess the translated version from a variety of text types Particularly, in

a study on assessing the Turkish – English translation, Çakır (2013 made a conclusion

on the translation quality based on translators’ comment and evaluation The results showed that despite there are basic text types, Google Translate still provide unstable performance due to lack of context knowledge Nevertheless, it is relatively evident that there has been no research papers or published articles about the use of Google Translate for travelling purpose What’s more, the past studies did not focus on the language pair Vietnamese – English but other language pairs, which leads to the researcher’s conclusion that this research paper can contribute to any related Vietnamese – English translation papers Finally, even though there are papers assessing the efficiency and limitations of Google Translate, it only addressed the perspectives and assessment from translation students or professional translators As this research paper focus on the customer, specifically middle-aged clients, the approach, and analysis will be different and offers a diverse aspect

In conclusion, after conducting in-depth research for definitions and concepts from previous findings, the researcher believes that there is room for doing this research paper as previous findings had not provided any real-life situations or the definitions and functions were not clarified and testified in reality Thus, this research paper strives for being one of the insightful paper about the efficiency of limitations

of Google Translate

CHAPTER 2: RESEARCH METHODOLOGY 2.1 Data collection instruments

To collect data for this paper, the researcher employs two types of data collection instruments, which are questionnaire (for quantitative data) and semi-structured interviews (for qualitative data)

2.1.1 Questionnaire

The researcher decided to use questionnaire as a means of collecting data for its merits In detail, the questionnaire is conducted in the open-ended and close-ended form with around 10-15 questions surrounding their experience with Google Translate when travelling The researcher hopes that participants can provide answers about the efficiencies and limitations that they found when using Google Translate Specifically, the questionnaire will include 2 main parts While the first part covers questions about the Google Translate using habit, the questions in the second part will try to elicit opinions to answer the research questions The questions will ask about the context when the participants had to use Google Translate, which function of Google Translate did the participants choose and ask for the participants’ opinions on the translations of different text types provided by Google Translate by using Likert-scale questions The questionnaire in paper will be directly given to the participants along with clear explanation from the researcher The use of questionnaire helps gathering a large amount of information conveniently, which the participants would only have to read and respond by ticking the options or writing short answers (Wallace, 1998) Since the research purpose is to assess the experience of the participants towards translations provided by Google Translate, the use of questionnaire, both in open and closed form, can receive demographic information and their answer to the questions Another reason of choosing this means of data collection is that it has

a simple and quick format for the participants to fill in, which is quite convenient for later analysis

2.1.2 Semi-structured interview

Besides conducting questionnaire, the researcher also opts for using interview to collect data, specifically semi-structured interview Expectedly, the interview will last for about 20-30 minutes to have more detailed information and opinion about the topic The semi-structure interview will dig deeper onto the issue by asking more detailed questions about the context, the translation process, and any further opinions As the researcher uses the experience for data analysis, using this method enables the researcher to elicit detailed information about personal feelings, perceptions and opinions and other questions that are not included in the interview to be answered Doing interview also assures that the research paper achieves higher response rate All questions will be open-ended to elicit answers from the participants The interviews will not be restricted as the questions only play the role as a guideline

2.2 Sampling method:

In terms of the sampling method for this paper, this research opt for simple random sampling method The researcher chose the simple random sampling method due to the fact that the population in this paper is middle-aged Vietnamese is too large; therefore, simple random sampling can enable the researcher to cull a smaller sample size from a larger population and use it to research Furthermore, in the aforementioned research objectives, it was addressed that the paper hopes to have some generations, which makes simple random sampling become a suitable match

By using simple random sampling, it “eliminates bias from the selection procedure and is associated with the minimum amount of sampling bias compared to other sampling methods.” (Gravetter, F.J & Forzano, L.B, 2011).Therefore, research findings resulting from the application of simple random sampling can be generalized due to representativeness of this sampling technique and a little relevance of bias

2.3 Scope of the study

To guarantee the highest confidence level, the research will be conducted by approaching a sample of 100 participants middle-aged Vietnamese They are all considered themselves as active tourists since they do travel frequently and greatly depend on this translation applications Their language proficiency is not fluent so they find it difficult to communicate with foreigners in English The participants have tried Google Translate and its functions before and are willing to share their opinions

towards Google Translate

2.4 Data collection procedure

2.4.1 Collecting data from questionnaires

Overall, the procedure will consist of three steps including:

Step 1: Piloting the questionnaires

In this stage, the researcher will conduct the questionnaires based on the research objectives and what information the research paper wants to yield The main purpose of this stage is to check any misunderstandings and whether the questions are relevant and relatable to the participants The number of questionnaires will not be too many and shall cover answers for the three research questions

Step 2: Distributing the questionnaires

After the piloting period, the researcher will distribute the questionnaire

In order to do this, the researcher must ask for the participants’ permission and willingness to let their information be studied During this process, it is essential to note that the researcher make clear and detailed instructions on filling the questionnaires for the participants

Step 3: Collecting and numbering the questionnaires

This is the last part of the procedure as the researcher will collect and keep track of the questionnaire response It is necessary for researcher to double check the responses for future use

2.4.2 Collecting data from semi-structured interviews

Step 1: Contacting the participants

After distributing questionnaires, for further understanding, the researcher will create a list of guideline questions and then select and contact participants via messages and direct calls Interview appointments will then be made under the agreement of both sides

Step 2: Conducting the interviews

As the researcher intends to have open-ended interviews, the interview will proceed following some guideline questions There will be spontaneous questions during the interview The interviews will first ask the participants about some general information and their impression about Google Translate

Step 3: Transcribing the interviews

Recording for interview will also be made and the researcher will have the responsibility to transcribe the interviews Additionally, the researcher will follow the notes being made for further analysis

CHAPTER 4: FINDINGS AND DISCUSSION

In the previous chapter, the methodology applied to conduct this research has been presented in detail To be specific, aspects namely the description and justification of the choices regarding the data collection instruments, participants as well as the procedure of collecting and analyzing data were put forward In this chapter, the results from the collected data with analysis based on the theoretical framework from the literature are presented to answer the research questions Subsequently, this chapter also provides a discussion of the results Overall, 84 participants have answered the questionnaire and 10 of them were asked to join the interview for further findings

4.1 Question 1: What are the main purposes of using Google Translate and what features are mostly used by middle-aged tourists?

In order to understand the behavior of middle-aged tourists using Google Translate, the researcher posed two questions to discover the purposes of middle-aged tourists when using Google Translate and which features they used in the questionnaire The researcher listed out some contexts that may need Google Translate and specific Google Translate features and asked the participants to choose the answers that best fit their experience

Figure 1: The main purposes that middle-aged tourists need Google Translate

What do you use Google Translate for? (You can choose more than 1)

Asking for directions

when traveling on roads

Asking for help when

going to the hospital

Overall, among the 5 purposes that had been listed out, there were 59 participants choosing “ordering food”, which accounts for 70.2% Other main purposes that this group of participants selected include “asking for directions when traveling on roads” (67,9%) and "asking for help when going

to the hospital"(69%) The two purposes which received the least votes were

“to bargain goods” and “purchasing tickets” with the relative percentage of 41,7% and 44% In addition, the researcher also yielded the same results from the interviewees as most of them agreed that they use Google Translate

to order food a lot

One possibility of this trend is that going to foreign countries might cause major changes in eating habits due to the cultural and culinary differences Therefore, middle-aged tourists may want to know what ingredients are included in the dish in order to avoid allergies or unwanted reactions Another possibility for people to use Google Translate when ordering food is that they may want to discover the food in the visiting countries Indeed, one participant shared that: “I have a lactose-intolerant issue so I should know the ingredients in advance before ordering food Also, as someone who likes cooking, I am also interested in getting to know the cooking methods and also the ingredients that do not exist in Vietnam.” Another participant also shared that as they are tourists, they sometimes visit street food vendors and

as they don't know the language, they need to use GT to assist them.”

Regarding the option “asking for help when going to the hospitals”, the results from both questionnaires and interviews reflect that it is plausible to

be the main purpose as they were middle-ages and may have physical or mental health problems One interviewee recounted the occasion when he had an accident while traveling with his peers and ended up going to a hospital for further check-ups It was Google Translate which he had to frequently use in order to understand the doctors’ guidance and instructions Some of the interviewees also have used Google Translate in the hospitals when dealing with their health issues

Even though there were mixed opinions concerning the need of translation tools for shopping, all of the interviewees agreed with the questionnaire participants that it is important to be equipped with Google Translate when going out as they are not used to directions written in English The underlying reason is to avoid getting lost or paying fine “Even though some road maps are similar to what we have seen in Vietnam, we always use Google Translate to make sure we get into the correct direction.”, a participant has shared

From the researcher’s perspective, it comes as no surprise that buying tickets was not a mostly-chosen as the use of non-verbal languages can be used In some cases, as the middle-aged tourists join tour travel, they do not even have to buy tickets “We often send someone who has basic communication skills in English to buy tickets for the whole group” “It is quite easy to buy tickets using sign languages since the staff at the information and ticket center understand and can correctly interpret our requests.” These opinions have backed up the results that buying tickets is

not an essential context to use Google Translate

Bargaining goods are the least context to be chosen for the use of Google Translate, which may be reasonable as people can use sign language such as pointing, nodding, and shaking heads to communicate This is further

supported by the participants in the interview by indicating that when going

shopping, it is unnecessary to use translation tools “You can negotiate prices

by using calculators and sign language” However, some interviewees have

contrasting opinions as they tend to use Google Translate to understand the

functions of specific products/items before buying,

Figure 2: Features of Google Translate that were used by participants

It can be seen that from the chart above, the most prominent features that

have been used by middle-aged participants are speech translation and image

translation The chart also discovers that between the two features, image

translation surprisingly becomes the more popular feature among

middle-aged users with 65 over 85 participants (equivalent to 76,5%) Mobile app

translation also received 46 votes while the participants do not prefer written

word translation and handwritten translation Only 36,5% of participants

choose written word translation and 31,85% choose handwritten translation

to use when going abroad The researcher also found out that among the

features of Google Translate, it is quite surprising that Image Translation,

Speech Translation, and Mobile App Translation are the top three most-used

features