Artificial Intelligence (AI) Chatbot as Language Learning
#WakeUp
😈💩👎
Medium: An inquiry
Nuria Haristiani
Japanese Language Education Department, Universitas Pendidikan Indonesia, Jl. Dr.
Setiabudhi 229, Bandung, Indonesia
Abstract. In facing industry revolution 4.0, utilizing advanced information and computer
technology in educational environment is crucial. One of the advanced computation
technologies that can be used for learning, especially language learning, is chatbot. Chatbot is
a computer program based on artificial intelligence that can carry out conversations through
audio or text. This study intends to find out and analyze the types of artificial intelligence in
the form of chatbots and the possibility of their use as language learning medium. The data in
this study obtained from literature review on chatbot researches, and from observation results
on chatbot-based language learning medium developed by the author. The results indicated that
chatbots have a high potential to be used as a language learning medium, both as tutor in
practicing language, and as independent learning medium. Moreover, research results revealed
that language learners are interested in using chatbots because they can be used anytime and
anywhere, and they are more confident in learning languages using chatbots than when dealing
directly with human tutors.
1. Introduction
The industrial revolution 4.0 has an impact on the urgency of education field to be able to keep up
with these developments, which later brought the term Education 4.0 [1, 2, 3]. In embodying
Education 4.0, one of the most required ability from educators and educational practitioners is to be
able to integrate modern technology in their teaching [3]. The rapid development of smart phone
technology, social media, and artificial intelligence (AI), provide challenges for educational
practitioners to utilize these technologies in developing advance learning media. In latest decades,
artificial intelligence utilization to develop applications is massively conducted, and its products used
in almost every aspects of our life. This type of communication which occurs through digital
technology rather than in person is called computer-mediated communication (CMC) [4]. CMC forms
including instant messaging, email, chat rooms, online forums, social networks, and chatbot or
chatterbot [5, 6]. Chatbot is a computer program or artificial intelligence which carries out
conversations through audio or text [7], and interact with users in a particular domain or topic by
giving intelligent responses in natural language [8, 9]. Chatbot for general purposes and for
educational purposes have been developed [4,10,11]. However, despite chatbots’ unlimited possibility
to enhance language teaching and learning, the concept of chatbot including its advantages as
language learning medium is not yet widely known. Therefore, the purpose of this study is to analyze
the types of artificial intelligence in the form of chatbots and the possibility of their use as language
learning medium. This study also aims to observe a chatbot constructed as Japanese language learning
medium developed by the author and team, and reports its’ results as an inquiry to find out further
about the possibility of chatbot to enhance language learning and teaching.
International Conference on Education, Science and Technology 2019
Journal of Physics: Conference Series 1387 (2019) 012020
IOP Publishing
doi:10.1088/1742-6596/1387/1/012020
2
2. Method
This study is a descriptive method study. The data collection in this study conducted through literature
reviews on previous researches about chatbot and its’ use. Literature reviews performed to identify the
type of chatbots, especially those developed for educational purposes including language learning, as
well as to identify their advantages/disadvantages in language teaching and learning. This study also
include a report on observation results on chatbot-based language learning medium developed by the
author and team, namely Gengobot.
3. Findings and Discussion
The basic mechanism of chatbot begins with the message sent by the user. The message then
processed by NLP (Natural Language Processing), and chatbot responded by replying to the message
according to the existing database (see figure 1). For example, when a user sent "how are you?"
message, chatbot will look for answers that match this question in the database such as “I am fine”,
“Great!” etc.
Figure 1.
The mechanism of
chatbot
.
3.1. The types of Chatbot
The type of chatbot found in this study can be categorized into three types based on its’ structure,
purpose, and audience. The sub-categories and their functions is concluded in table 1.
Table 1.
T
ypes of
chatbot
.
Category
Sub
-
category
Function
Structure Flow chatbot A tree-based chatbot. This chatbot has fixed responds set by the developer,
and only responds to questions that are already in the database. Flow chatbots
include buttons, keywords, and catchphrases instead of free writing to drive
the client down
the predefined path.
Artificially
intelligent
Chatbot with artificial intelligence has the ability update their knowledge and
perception from previous conversations and users’ experience, letting the
users engage more freely.
Hybrid This type of chatbot combines the concepts of Flow and AI chatbots. This
chatbot can understand and communicate with users, but remains in the
pattern determined by the developer.
Purpose Functionality This chatbots have certain functions depends on the developer (i.e. chatbot for
learning, personal assistant, reminder, online shop assistant, etc.)
Fun
Chatbot that intended only for entertainment (i.e. games,
funbot
, etc.).
Audience Generalist This chatbot has general knowledge that we can ask directly. I.e. Siri
developed by Apple ,and Cortana developed by Microsoft. Both Chatbots can
help us solve common problems such as searching for restaurants, locations
and more.
Specialist This chatbot focus on one constrained thing and do that one thing extremely
well
(i.e. chatbots that used t
o serve customers online when ordering items).
International Conference on Education, Science and Technology 2019
Journal of Physics: Conference Series 1387 (2019) 012020
IOP Publishing
doi:10.1088/1742-6596/1387/1/012020
3
Table 1 shows that chatbot has several categories and can be developed according to the developer’s
necessity. However, chatbots that are developed for educational purposes tend to use artificially
intelligence structure. Artificially intelligent chatbot for general purpose such as MILABOT has also
been developed [10]. MILABOT is a deep reinforcement learning chatbot which capable of
conversing with humans on popular small talk topics through both speech and text. The system
consists of an ensemble of natural language generation and retrieval models, including template-based
models, bag-of-words models, sequence-to-sequence neural network and latent variable neural
network models [10]. Chatbots that developed for educational purposes, especially for language
learning, are described further in the following sub-section.
3.2. Chatbot in Language Learning and Teaching
Chatbots development to utilize learning and teaching have been conducted. Freudbot has been
developed for psychology students to find out about student-content interaction in distance education.
The results shows that the basic analysis of the chatlogs indicated a high proportion of on-task
behavior. The findings also suggests that chatbot technology may be promising as a teaching and
learning tool in distance and online education [12]. Chatbot use is also compared with humanoid robot
in science lecture class, and reported that the visualization using chatbot was helpful for students to
understand the lecture smoothly [13]. However, researches on chatbot use and development to
enhance language learning rather difficult to find. This study identified researches on language
teaching and learning as seen in table 2.
Table 2.
Chatbot researches on language learning.
First author
(year)
Chatbot name Subject Focus Sample Research type
Jia (2004) - English,
Germany
as Foreign
Language
Application of a Web-
based Chatbot system
on foreign language
teaching
1256 Experiment
Fryer (2006) Cleverbot English as
Foreign
Language
Chatbot as English
language learning tools
211 Experiment &
survey
Jia (2009) CSIEC chatbot English
Learning
A computer assisted
English learning chatbot
based on textual
knowledge and
reasoning
1783
Experiment, survey
& questionnaire
Goda (2014) Cleverbot English as
Foreign
Language
The use of Chatbot
before online EFL
discussion and The
effect on critical
thinking
130 Comparison based
(Experimental &
Control group)
Fryer (2017) Cleverbot English as
Foreign
Language
Comparison of chatbot
and human task partners
in English learning
122 Comparison based
(Pre-test & Post-
test)
Table 2 shows that chatbot researches were mainly found in English language learning
[14,15,16,17,18]. Research reported that the dialogs using chatbot are mostly very short because the
users find the computer is much less intelligent as a human, since the responses from the computer are
often repeated and irrelevant with the topics and the context. However, the results also indicates that
International Conference on Education, Science and Technology 2019
Journal of Physics: Conference Series 1387 (2019) 012020
IOP Publishing
doi:10.1088/1742-6596/1387/1/012020
4
many participants are very interested in using chatbot as chatting partner in speaking foreign language,
since it is accessible anywhere and anytime, while it is not easy to find native speakers as human
chatting partner. The learners also more confident communicating with chatbot which is obviously
less intelligent as the human themselves. It would be pedagogically attractive for the learner to chat
with a system of artificial intelligence which could “really” understand the natural language and
communicatively generate the natural language to form a human-like dialog [14].
Further, the main chatbot used in language learning researches as seen in Table 2 is Cleverbot, which
developed by British AI scientist Rollo Carpenter in 1986 and went online on 1997. The results of
researches of using Cleverbot reported that most students enjoyed using this chatbot [15,17]. They
also generally felt more comfortable conversing with the bots than a student partner or teacher.
However, the results also suggest that chatbots are generally only useful for advanced and/or very
keen language students. Language teachers also need to get involved and bring chatbot technology
into the foreign language learning classroom as a permanent tool for language practice [15]. Research
reported that preceding conversation before classroom discussion with a chatbot lead to an increase in
the number of contributions that students made to discussion [17]. Moreover, pre-discussion with a
chatbot also could increase the students’ awareness of critical thinking and enable them to form
inquiring mindsets [17]. However, the result of comparisons in speaking task with chatbot and human
partner indicated a significant drop in students' task interest with chatbot, but not human partner. The
reason of drop in task interest with chatbot was caused by novelty effect [18]. On the other hand,
CSIEC chatbot reported successfully helped students with course unit review, make the students more
confident, and improved students’ listening ability, as well as enhanced students’ interest in language
learning. The comparison of examination results before and after the using chatbot showed great
improvement of students’ performance [15].
From above results, it is understood that the use of chatbot gave many advantages in language learning
and teaching, as in enhancing classroom motivation and learning [15,19]. However, chatbot also
reported to have flaws comparing to human partner, especially in the novelty aspect [18]. Several
researches and chatbot development for English language learning have been conducted, but chatbot
development and researches in other languages teaching and learning is still difficult to find. As an
attempt to answer this challenge, the author and team tried to develop a chatbot-based multi-language
grammar application, namely Gengobot, which will be introduced further in the next sub-section.
3.3. Gengobot as Japanese Language Learning Medium
Gengobot is a chatbot-based dictionary application about multi-language grammar developed by the
author, using CodeIgniter (CI) framework. CI is a PHP framework that can be used to develop PHP-
based website application without the necessity to write all the code from the beginning. CI framework
was chosen because it is an opensource framework and free to modify, smaller than other framework,
and uses MVC (Model-View-Controller) concept that functioned in programming process to call
needed databases easily. The main purpose of Gengobot development is to provide a Japanese
grammar learning medium for beginner level of Japanese language learners. However, to broaden its’
use, the application also equipped with grammar contents in English and Indonesian, and integrated
with social media LINE. LINE official account has Messaging API feature that allows an account to
run chatbot that has been created (see figure 2). The database system used in Gengobot is MySQL (see
figure 3 and figure 4). MySQL was chosen because it is free licensed, and the database structure used
in MySQL is in table form which is flexible and easy to use. The database created for this chatbot
including: (1) Database of user data storage, including name, language, training score, etc.; (2)
Grammar database in three languages; (2) Questions database and their answer (for ‘Exercise’
feature).
International Conference on Education, Science and Technology 2019
Journal of Physics: Conference Series 1387 (2019) 012020
IOP Publishing
doi:10.1088/1742-6596/1387/1/012020
5
Figure 2. LINE Messaging API and chatbot
mechanism.
Figure 3. Gengobot database creation using
MySQL
.
Figure 4. The process of importing database using
MySQL.
Figure 5. Gengobot interface designing process.
Gengobot consists of fixed responds including buttons, keywords, and catchphrases set as database.
The buttons includes several menus as ‘Menu’, ‘Help’, ‘Language’, and ‘Contact’ option as shown in
figure 5, as well as several sub-menus that developed under each menu. From observation and
evaluation results, Gengobot was successfully developed as a language learning medium, especially to
support students’ learning about Japanese grammar with description available in English and
Indonesian. Gengobot also considered user friendly since it could be accessed through social media
LINE. The features in Gengobot also includes ‘Exercise’, so the students not only able to learn about
Japanese grammars, but also able to test their knowledge about grammars they have learned. However,
Gengobot is a flow chatbot, so the interaction between users/students and chatbot still very limited to
the inputted database. Although as grammar learning application it is still considered sufficient, to
improve its’ use for enhancing language learning, as well as other language skills teaching and
learning, Gengobot still need to be developed further in its’ technology and features.
4. Conclusion
This study aimed to analyze the types of chatbots and the possibility of their use as language learning
medium. From the results, it is known that chatbot can be categorized into three types, and has
advantages and disadvantages. As the advantages, chatbot is reported can help language learners
through six ways: (1) students tend to feel more relaxed talking to a computer than to a person; (2)
chatbots are willing to repeat the same material with students endlessly; (3) many bots provide both
text and synthesized speech, allowing students to practice both listening and reading skills; (4) Bots
are new and interesting to students; (5) students have an opportunity to use a variety of language
structures and vocabulary that they ordinarily would not have a chance to use; (6) chatbots could
potentially provide quick and effective feedback for students’ spelling and grammar [15]. However,
chatbot also reported to have a flaw on its novelty aspects and need to be improved. This study also
observed a chatbot-based Japanese language learning medium developed by the author, namely
Gengobot. As the results, Gengobot have a high potential to be used as a Japanese language learning
medium especially in learning grammar, yet need to be developed further in its’ technology and
features.
International Conference on Education, Science and Technology 2019
Journal of Physics: Conference Series 1387 (2019) 012020
IOP Publishing
doi:10.1088/1742-6596/1387/1/012020
6
5. References
[1] Harkins A M 2008 Leapfrog principles and practices: Core components of education 3.0 and 4.0.
Futures Research Quarterly 24(1) pp 19-31
[2] Puncreobutr V 2016 Education 4.0: new challenge of learning St. Theresa Journal of
Humanities and Social Sciences 2(2)
[3] Hussin A A 2018 Education 4.0 Made Simple: Ideas For Teaching International Journal of
Education and Literacy Studies 6(3) pp 92-8
[4] Hill J Ford W R and Farreras I G 2015 Real conversations with artificial intelligence: A
comparison between human–human online conversations and human–chatbot conversations.
Computers in Human Behavior 49 pp 245-50
[5] Tagliamonte S A and Denis D 2008 Linguistic ruin? LOL! Instant messaging and teen language
American speech 83(1) pp 3-4
[6] Thurlow C L Lengel L L and Tomic A 2004 Computer Mediated Communication: Social
Interaction and the Internet
[7] Shevat A 2017 Designing bots: creating conversational experiences O'Reilly Media, Inc
[8] Abdul-Kader S A and Woods J C 2015 Survey on chatbot design techniques in speech
conversation systems International Journal of Advanced Computer Science and Applications
6(7)
[9] Azwary F Indriani F and Nugrahadi D T 2016 Question Answering System Berbasis Artificial
Intelligence Markup Language Sebagai Media Informasi KLIK-KUMPULAN JURNAL
ILMU KOMPUTER 3(1) pp 48-60
[10] Serban I V Sankar C Germain M Zhang S Lin Z Subramanian S Kim T Pieper M Chandar S Ke
N R and Rajeshwar S 2017 A deep reinforcement learning chatbot. arXiv preprint
arXiv:1709.02349
[11] Chen J A Tutwiler M S Metcalf S J Kamarainen A Grotzer T and Dede C 2016 A multi-user
virtual environment to support students' self-efficacy and interest in science: A latent growth
model analysis Learning and Instruction 41 pp 11-22
[12] Heller B Proctor M Mah D Jewell L and Cheung B 2005 Freudbot: An investigation of chatbot
technology in distance education InEdMedia+ Innovate Learning pp 3913-3918
[13] Matsuura S and Ishimura R 2017 Chatbot and dialogue demonstration with a humanoid robot in
the lecture class. International Conference on Universal Access in Human-Computer
Interaction pp 233-246
[14] Jia J 2004 The study of the application of a web-based chatbot system on the teaching of foreign
languages. Society for Information Technology & Teacher Education International
Conference pp 1201-1207
[15] Fryer L and Carpenter R 2006 Bots as language learning tools Language Learning &
Technology 10(3) pp 8-14
[16] Jia J 2009 CSIEC: A computer assisted English learning chatbot based on textual knowledge
and reasoning Knowledge-Based Systems 22(4) pp 249-55
[17] Goda Y Yamada M Matsukawa H Hata K and Yasunami S 2014 Conversation with a chatbot
before an online EFL group discussion and the effects on critical thinking. The Journal of
Information and Systems in Education 13(1) pp 1-7
[18] Fryer L K Ainley M Thompson A Gibson A and Sherlock Z 2017 Stimulating and sustaining
interest in a language course: An experimental comparison of Chatbot and Human task
partners Computers in Human Behavior 75 pp 461-8
[19] Coniam D 2008 Evaluating the language resources of chatbots for their potential in English as a
second language ReCALL 20(1) pp 98-116
