INTERNATIONAL CONFERENCE
Impacts of Globalization on Quality in Higher Education
SEAMEO RETRAC - Ho Chi Minh City
HOW CAN GOOGLE TRANSLATION MACHINE (GTM)
ASSIST VIETNAMESE LEARNERS OF ENGLISH?
A CASE STUDY OF TRANSLATING INTERROGATIVE
SENTENCES AND SOME SUGGESTIONS FOR
IMPROVEMENT
Presenter: Dr. Nguyen Thi Chau Anh
1
1. INTRODUCTION
1.1. Rationale
1.2. Objectives
1.3. Methodology
1.4. Significance of the study
1.5. Scope of the study
1.6. Overview
2
a. Research questions:




How efficient and/or deficient are the
target language texts produced by Google
Translate Service?
What are the most common problems
that characterize that translation service and
how to solve them for improvement?
How does GTM assist students in learning
English?
The present research is an attempt to find
answers to these questions.
3
b. Hypotheses of the research
One would expect major problems:
Firsly, on the semantic level and in particularly on
ambiguity from polysemy;
Secondly, errors on some modal verbs, particles (à,
ư, nhỉ, nhé, nha, phải không, hả, chứ?), modality
makers, sentence operators in Vietnamese
interrogatives.
It is expected that GTM program has been mainly
designed to solve the problems in general.
But what has the practical experience actually
revealed? That is the basic concern of this paper.
4
A CASE STUDY OF TRANSLATING
INTERROGATIVE SENTENCES
2.1 Theoretical and practical background
2.1.1 What is polysemy?
 The term polysemy is used in linguistics as a
means of categorizing and studying various
aspects of languages. The opposite of a
polysemy is a heterosemy, which means the
word has only a single meaning.
 Polysemy refers to a word that has two or
more similar meanings
5
2.1.2. How GTM translate in its service on line?
Approaches and application
Bernard Vauquois'
pyramid showing
comparative depths of
intermediary
representation,
interlingual machine
translation at the peak,
followed by transferbased, then direct
translation.
6
THEORETICAL BACKGROUND


These methods require extensive
lexicons with morphological, syntactic,
and semantic information, and large
sets of rules.
Machine translation can use a method
based on dictionary entries, which
means that the words will be translated
as they are by a dictionary.
7
2.2 Describes and explains the case study and proposes the
rules and technique called “input code” for
disambiguating words, which makes GTM more reliable
Dealing with interrogatives, V – E
translation by Google Translation
Machine (GTM) is certainly the case that
has to be carefully examined and
addressed for the improvement of the
GTM
8
Examples: modal particle for questions
Firstly, the 6 accent tones in Vietnamese were
not understood by GTM and often led to
ambiguity and made lexical errors. It was
rather impossible for the GTM program to
find equivalents to source language (SL)
ambiguous items.
nha? – nhà (home)
nhỉ ? - nhi (children)
9
Some common words for 8x/9x in Vietnamese language
and Proper nouns (which do not exist in the corpus/
dictionaries) can be used in disambiguating for the case of
polysemy
Examples: Replacing these words and put them in
the parentheses ( ):
1.
hông, không --- ko, hok
2.
à, ạ --ak
3.
hả, ha --hak
4.
gì ---ji
5.
nhỉ --nhik
6.
nha --nhak
10
The length of interrogative sentences in Vietnamese in
comparison with that in English (N= 320)
Distribution of Frequency in English
Distribution of Frequency in Vietnamese
100
80
80
60
60
40
40
20
Frequency
20
Std. Dev = 18.88
Std. Dev = 20.68
Mean = 32.3
Mean = 36.8
N = 320.00
0
0.0
20.0
10.0
40.0
30.0
60.0
50.0
80.0
70.0
Distribution of Frequency in English
100.0
90.0
120.0
110.0
130.0
N = 320.00
0
0.0
20.0
10.0
40.0
30.0
60.0
50.0
80.0
70.0
100.0
90.0
120.0
110.0
130.0
Distribution of Frequency in Vietnamese
11
EXAMPLES
Đi ăn cơm không?
Google translates: Take no rice to eat? (wrong)
a)
Bạn có thể cho tôi xin ý kiến của bạn về
vấn đề này (dc_ko_ak)?
Google translates:
a)
a) Can you give me your opinion on this
issue (dc_ko_ak)?
a)
12
Hypothesis for the case study of experiment

“If we follow the instructions for
Vietnamese language input the
results will be better and GTM can
be improved by enhancing the
quality of reliability in V- E
translation”
13
The first result (N=108)
(the interrogative sentences from language
materials and text books) with the
15.74
Test 1
84.26
right
sentences
%
74.07
Test 2
wrong
sentences
25.93
%
0
10
20
30
40
50
60
70
80
90
14
The second result (N=36)
(The interrogatives from students’ use)
120
5.56
100
%
Wrong sentences
80
60
100
94.44
40
%
Right sentences
20
0
Test 1
Test 2
15
3. HOW CAN GOOGLE TRANSLATION
MACHINE (GTM) ASSIST VIETNAMESE
LEARNERS OF ENGLISH?

Need analysis of using GTM in the
perspective of making and translating
interrogatives for English study
How might this learning aid can be used in
social interaction systems and products so
that users maximize what they learn on the
Web?
16
Gagne’s five categories of human
performance established by learning.





Intellectual skills (“knowing how” or having
procedural knowledge)
Verbal information (being able to state ideas,
“knowing that”, or having declarative knowledge)
Cognitive strategies (having certain techniques of
thinking, ways of analyzing problems and having
approaches to solving problems)
Motor skills (executing movements in a number of
organized motor acts such as playing sports or
driving a car)
Attitudes (mental states that influence the choices
of personal actions)
17
The nine events for students’ actions and
performance in class are as follows:

Giving students a task to do as home assignments
individually or in group work. Students have to ask
questions in different purposes beside getting
information for details to get an interesting short
story or an event (imaginary or true story happened
somewhere else) told by one classmate in class, by
watching video clips in you tube (download from
4share), or by chatting in Skype, Facebook or in
Yahoo Messenger with their teachers, another
classmate or an advisor; and save the data collection
as lessons in Drop Box.
18
1.
2.
3.
4.
5.
Gaining Attention: Ss listen or read one story as an
example from the teacher’s folder in the Sky drive;
Informing Learners of the Objective: Ss have to
understand it and retell it in more details by means of
media and send it to their sky drive for discussions
and sharing ideas;
Stimulating Recall of Prior Learning: Ss prepare the
questions to ask by translating Vietnamese questions
to English performed by GTM and edited by students
themselves;
Presenting the Stimulus: Ss will get the answers for
their questions and high marks or small presents for
asking enough correct and suitable questions as
required;
Providing Learning Guidance: Ss will receive their
teacher’s feedback as soon as they send the questions
to sky drive to the teacher;
19
Student assignments for using
questions with GTM
6.
7.
8.
9.
Eliciting Performance: Ss will be shown how
to make questions correctly with GTM;
Providing Feedback: Ss will be shown and
instructed the way to make good questions
for the story;
Assessing Performance: Ss use the answers
they get to retell that story in detail with
joys, and relaxations;
Enhancing Retention and Transfer: Ss will
have some interesting stories made from
their own information by asking questions.
20
4. CONCLUSION



The GTM program can be described
with the techiques “input code” as
Encoding the meaning of the source
text; and
Re-encoding and decoding this
meaning in the target language.
21
The researcher’s conclusions
Based on the results of trial tests, These conclusions
totally support the hypothesis of the study
GTM still maintains some advantages.
First, machine translation is much faster than human
translation. Second, machine translation has a much
huger quantity of vocabulary than human and has its
useful tool of how topronounce these words in each
question.
Hopefully, it can be capable of translating interrogatives
more reliable and translating articles according to
different types effectively in the near future.
22
Google believes that translating one source
language to another target one in our future
is crucial for everyone. We look forward to
jointly working with these researchers,
experts and linguists concerning GTM
program and hope that we will push the
frontier of social interactions research with
GTM to the next level.
23
TÀI LIỆU THAM KHẢO










1. http://pewinternet.org/topics/Future of the internet.aspx and
http://imaginingtheinternet.org.
Janna Quitney Anderson, Elon University - Lee Rainie, Pew Internet & American Life Project
(2010), The Future of the Internet, Pew Research Center’s Internet & American Life Project
An initiative of the Pew Research Center,
2. www.ccsenet.org/ells English Language and Literature Studies, Vol. 2, No. 1; March
2012.
3. www.ccsenet.org/ells, English Language and Literature Studies, Vol. 2, No. 1; March
2012.
4. a. http://en.wikipedia.org/wiki/Machine_translation: From Wikipedia, the free
encyclopedia.
b. http://www.google.com.vn.
5. http://www.icels-educators-for learning.ca/index.php?option=com_content&view=article
& id=54&Itemid=73
Robert Gagne, Robert Gagne’s Five Categories of Learning Outcomes and the Nine Events
of Instruction*, Description, Main Elements, Table of the 9 Events of Instruction, References
6. http://www.wisegeek.com/what-is-a-polysemy.htm
24
THANK YOU FOR ATTENTION!
?
25
THE END
?
26