2023, Volume 26, Issue 1

Special Issue on "Human-centered AI in Education: Augment Human Intelligence with Machine Intelligence"

Guest Editor(s): Stephen J.H. Yang, Hiroaki Ogata and Tatsunori Matsui

Download Table of Content in PDF

Download Complete Issue in PDF

Full Length Articles

Zhi Liu

National Engineering Research Center for Educational Big Data, Faculty of Artificial Intelligence in Education, Central China Normal University, China // zhiliu@mail.ccnu.edu.cn

Ning Zhang

National Engineering Research Center for E-Learning, Faculty of Artificial Intelligence in Education, Central China Normal University, China // zhangning97@mails.ccnu.edu.cn

Xian Peng

National Engineering Research Center for Educational Big Data, Faculty of Artificial Intelligence in Education, Central China Normal University, China // pengxian@ccnu.edu.cn

Sannyuya Liu

National Engineering Research Center for Educational Big Data, Faculty of Artificial Intelligence in Education, Central China Normal University, China // National Engineering Research Center for E-Learning, Faculty of Artificial Intelligence in Education, Central China Normal University, China // lsy.nercel@gmail.com

Zongkai Yang

National Engineering Research Center for Educational Big Data, Faculty of Artificial Intelligence in Education, Central China Normal University, China // National Engineering Research Center for E-Learning, Faculty of Artificial Intelligence in Education, Central China Normal University, China // 13659885363@163.com


ABSTRACT:

Grounded on constructivism, mining a complex mix of social and cognitive interrelations is key to understanding collaborative discussion in online learning. A single examination of one of these factors tends to overlook the impact of the other factor on learning. In this paper, we innovatively constructed a social-cognitive engagement setting to jointly characterize social and cognitive aspects. In the online discussion forum, this study jointly characterized students’ social and cognitive aspects to investigate interactive patterns of different social-cognitive engagements and social-cognitive engagement evolution across four periods (i.e., creation, growth, maturity, and death). Multi-methods including social network analysis, content analysis, epistemic network analysis, and statistical analysis was applied in this study. The results showed that the interactive patterns of social-cognitive engagement were affected by both social network position and cognitive level. In particular, students’ social network position was a vital indicator for the contributions to cognitive level of students, and cognitive level affected the related interactions to some extent. In addition, this study found a nonlinear evolutionary development of students’ social-cognitive engagement. Furthermore, maturity is a critical period on which teachers should focus, as the co-occurrence of social-cognitive engagement reaches a maximum level in this period. Based on the results, this multi-perspective analysis including social and cognitive aspects can provide insightful methodological implications and practical suggestions for teachers in conducting in-depth interactive discussions.


Keywords:

Social-cognitive engagement, Integrated analysis, Social network analysis (SNA), Epistemic network analysis (ENA), Knowledge building

Siu-Cheung Kong

Department of Mathematics and Information Technology, The Education University of Hong Kong, Hong Kong SAR // Centre for Learning, Teaching and Technology, The Education University of Hong Kong, Hong Kong SAR // siucheungkong@gmail.com

William Man-Yin Cheung

Centre for Learning, Teaching and Technology, The Education University of Hong Kong, Hong Kong SAR // williamcheung@eduhk.hk

Guo Zhang

Centre for Learning, Teaching and Technology, The Education University of Hong Kong, Hong Kong SAR // gzhang@friends.eduhk.hk


ABSTRACT:

Emerging research is highlighting the importance of fostering artificial intelligence (AI) literacy among educated citizens of diverse academic backgrounds. However, what to include in such literacy programmes and how to teach literacy is still under-explored. To fill this gap, this study designed and evaluated an AI literacy programme based on a multi-dimensional conceptual framework, which developed participants’ conceptual understanding, literacy, empowerment and ethical awareness. It emphasised conceptual building, highlighted project work in application development and initiated teaching ethics through application development. Thirty-six university students with diverse academic backgrounds joined and completed this programme, which included 7 hours on machine learning, 9 hours on deep learning and 14 hours on application development. Together with the project work, the results of the tests, surveys and reflective writings completed before and after these courses indicate that the programme successfully enhanced participants’ conceptual understanding, literacy, empowerment and ethical awareness. The programme will be extended to include more participants, such as senior secondary school students and the general public. This study initiates a pathway to lower the barrier to entry for AI literacy and addresses a public need. It can guide and inspire future empirical and design research on fostering AI literacy among educated citizens of diverse backgrounds.


Keywords:

Application development, Artificial intelligence literacy, Conceptual framework, Ethical awareness, University students

Yin Yang

Department of Mathematics and Information Technology, The Education University of Hong Kong, Hong Kong // yyin@s.eduhk.hk

Yun Wen

National Institute of Education, Nanyang Technological University, Singapore // yun.wen@nie.edu.sg

Yanjie Song

Department of Mathematics and Information Technology, The Education University of Hong Kong, Hong Kong // ysong@eduhk.hk


ABSTRACT:

The role of self-regulated learning in language learning has been widely acknowledged, and there is a growing number of studies on technology-enhanced self-regulated language learning (SRLL). This systematic review aims to provide a holistic picture of existing studies in this area by identifying the characteristics of published studies, the research methods used to evaluate SRLL effectiveness and the role of technology in SRLL. The review covered 34 empirical studies focusing on SRLL that were published from 2011 to 2020. The results showed varied characteristics of technology-enhanced SRLL studies, dominance of the use of quantitative methods, greater focus on examining students’ SRLL outcomes instead of their processes, and the role of technology in supporting the performance phase of students’ SRLL instead of the entire SRLL process. These findings have implications for using technologies to facilitate and examine the holistic process of students’ SRLL.


Keywords:

Systematic literature review, Technology, Self-regulated language learning (SRLL)

Wiwit Ratnasari

Department of Information Management, National Taiwan University of Science and Technology, Taipei, Taiwan, R.O.C. // ratnasariwiwit@gmail.com

Tzu-Chuan Chou

Department of Information Management, National Taiwan University of Science and Technology, Taipei, Taiwan, R.O.C // tcchou@mail.ntust.edu.tw

Chen-Hao Huang

Graduate Institute of Technology Management, National Taiwan University of Science and Technology, Taipei, Taiwan, R.O.C. // chhuang@mail.ntust.edu.tw


ABSTRACT:

The digital revolution has heavily influenced digital game-based learning, yet as the revolution progresses, the conception of such learning has shifted along with the increasing complexity of the digital environment. Our study thus aims to identify research standing at this important juncture and to explain the shift in digital game-based learning research fields by adopting an integrated approach of main path analysis that yields this topic’s knowledge diffusion. Using key-route 8 to construct the path, we collect a total of 2156 articles and their data from The Web of Science database. From over 30 years of digital game-based learning development, 26 of the most influential studies are identified and visualized using Pajek software. The findings show two development phases for this field: exploring the role of gaming for educational purpose as well as facilitating learning performance. The research focus in the first phase prominently explores the potentials of digital games for educational purposes, and then the focus evolves in the second phase into actualizing the identified potentials. We propose a framework of digital game-based learning affordance actualization to explain these shifting phenomena in the specific research fields. Furthermore, unveiling the changing conception of digital game-based learning research is important for instructional designers, scholars, and educators to truly understand how technology can enhance teaching and facilitate learning performance.


Keywords:

Affordance actualization, Main path analysis, Digital game-based learning, Citation network, Knowledge diffusion

Yu-chu Yeh

Institute of Teacher Education, National Chengchi University, Taipei City, Taiwan // Research Center for Mind, Brain & Learning, National Chengchi University, Taipei City, Taiwan // ycyeh@nccu.edu.tw

Yu-Shan Ting

Department of Education, National Chengchi University, Taipei City, Taiwan // pnm40275@gmail.com

Jui-Ling Chiang

Institute of Teacher Education, National Chengchi University, Taipei City, Taiwan // rayechiang@gmail.com


ABSTRACT:

Creativity mindset (CM), grit, and self-determination have been defined as critical motivational variables affecting learners’ self-efficacy. Therefore, this study pioneers the examination of the relationship between these motivational variables and creativity self-efficacy (CSE) during game-based learning. A Creativity Mindset Inventory (CMI) and a game-based learning intervention were employed. Participants for developing the CMI were 281 3rd to 6th graders, and those for the intervention were 114 3rd and 4th graders. The result revealed that the CMI included four constructs (growth-internal control, growth-external control, fixed-internal control, and fixed-external control). Moreover, the employed intervention enhanced the children’s growth CM and CSE. Regression analysis results suggest that self-determination mediates the influence of growth CM and grit on CSE. Additionally, growth CM, especially the growth-internal control CM, is a powerful predictor of self-determination and CSE. In contrast, fixed CM (the overall fixed CM, the fixed-internal control CM, or the fixed-external control CM) does not have any significant influence on self-determination or CSE. Notably, the findings of this study support that growth CM can be enhanced through a well scaffolded educational game. This study contributes to the field of game-based learning by developing a CM inventory, demonstrating a growth CM intervention, and clarifying influential factors to CSE during game-based training. While game-based learning has become popular among elementary school students, the findings of this study provide important insights into the design of game-based learning and creativity training.


Keywords:

Creativity, Game-based learning, Growth mindset, Grit, Self-determination

Semiotic Alternations with the Yupana Inca Tawa Pukllay in the Gamified Learning of Numbers at a Rural Peruvian School

Rosario Guzman-Jimenez, Dhavit-Prem, Alvaro Saldívar and Alejandro Escotto-Córdova

Rosario Guzman-Jimenez

Universidad de Lima, Perú // rguzman@ulima.edu.pe

Dhavit-Prem

Asociación Yupanki, Perú // dhavitprem@gmail.com

Alvaro Saldívar

Asociación Yupanki, Perú // yachay@yupanainka.com

Alejandro Escotto-Córdova

UNAM-Zaragoza, México // aescotto@unam.mx


ABSTRACT:

Yupana Inca Tawa Pukllay (YITP) is a ludic didactic resource based on semiotic alternation that, using the reading of numbers in the Inca numeral system, improves its equivalent Indo-Arabic reading. Twelve children from first to fourth grade of a bilingual (Spanish-Quechua), multi-grade elementary school in a small rural Peruvian community were assigned an electronic tablet with YITP and learned autonomously, without teachers during the COVID-19 pandemic. The results obtained show that: (a) they learned in a very short period of time (14 min - 05h 41 min) (b) they improved digit reading accuracy on the first attempt (c) they improved digit reading speed d) they achieved a high percentage of correct reading of numbers containing at least one zero digit. The results suggest the potential of YITP as an educational tool in the teaching-learning process of arithmetic.


Keywords:

Semiotic alternations, Yupana Inca Tawa Pukllay, Ethnomathematics gamification, Numeral system, Zero

Editorial

Human-centered AI in Education: Augment Human Intelligence with Machine Intelligence

Stephen J.H. Yang, Hiroaki Ogata and Tatsunori Matsui

Special Issue Articles

Nabeel Gillani, Rebecca Eynon, Catherine Chiabaut and Kelsey Finkel

Nabeel Gillani

Massachusetts Institute of Technology, USA // ngillani@mit.edu

Rebecca Eynon

University of Oxford, UK // rebecca.eynon@oii.ox.ac.uk

Catherine Chiabaut

The Robertson Foundation, USA // catherine.chiabaut@robertson.org

Kelsey Finkel

The Robertson Foundation, USA // kelsey.finkel@robertson.org


ABSTRACT:

Recent advances in Artificial Intelligence (AI) have sparked renewed interest in its potential to improve education. However, AI is a loose umbrella term that refers to a collection of methods, capabilities, and limitations—many of which are often not explicitly articulated by researchers, education technology companies, or other AI developers. In this paper, we seek to clarify what “AI” is and the potential it holds to both advance and hamper educational opportunities that may improve the human condition. We offer a basic introduction to different methods and philosophies underpinning AI, discuss recent advances, explore applications to education, and highlight key limitations and risks. We conclude with a set of questions that educationalists may ask as they encounter AI in their research and practice. Our hope is to make often jargon-laden terms and concepts accessible, so that all are equipped to understand, interrogate, and ultimately shape the development of human-centered AI in education.


Keywords:

K-12 education, Artificial intelligence in education, Educational data mining, Learning analytics, Natural language processing

Xinyi Huang, Di Zou, Gary Cheng, Xieling Chen and Haoran Xie

Xinyi Huang

Department of Mathematics and Information Technology, The Education University of Hong Kong, Hong Kong // hxinyi@eduhk.hk

Di Zou

Department of English Language Education, The Education University of Hong Kong, Hong Kong // dizoudaisy@gmail.com

Gary Cheng

Department of Mathematics and Information Technology, The Education University of Hong Kong, Hong Kong // chengks@eduhk.hk

Xieling Chen

Department of Mathematics and Information Technology, The Education University of Hong Kong, Hong Kong // xielingchen0708@gmail.com

Haoran Xie

Department of Computing and Decision Sciences, Lingnan University, Hong Kong SAR // hrxie2@gmail.com


ABSTRACT:

Artificial Intelligence (AI) plays an increasingly important role in language education; however, the trends, research issues, and applications of AI in language learning remain largely under-investigated. Accordingly, the present paper, using bibliometric analysis, investigates these issues via a review of 516 papers published between 2000 and 2019, focusing on how AI was integrated into language education. Findings revealed that the frequency of studies on AI-enhanced language education increased over the period. The USA and Arizona State University were the most active country and institution, respectively. The 10 most popular topics were: (1) automated writing evaluation; (2) intelligent tutoring systems (ITS) for reading and writing; (3) automated error detection; (4) computer-mediated communication; (5) personalized systems for language learning; (6) natural language and vocabulary learning; (7) web resources and web-based systems for language learning; (8) ITS for writing in English for specific purposes; (9) intelligent tutoring and assessment systems for pronunciation and speech training; and (10) affective states and emotions. The results also indicated that AI was frequently used to assist students in learning writing, reading, vocabulary, grammar, speaking, and listening. Natural language processing, automated speech recognition, and learner profiling were commonly applied to develop automated writing evaluation, personalized learning, and intelligent tutoring systems.


Keywords:

Artificial Intelligence, Language education, Bibliometric analysis, Automated writing evaluation, Intelligent Tutoring System

Fuzheng Zhao

Kobe University, Japan // Jilin University, China // zhaofz635@gmail.com

Gi-Zen Liu

National Cheng Kung University, Taiwan // gizen@mail.ncku.edu.tw

Juan Zhou

Tokyo Institute of Technology, Japan // juan.z.kt@gmail.com

Chengjiu Yin

Kobe University, Japan // yin@lion.kobe-u.ac.jp


ABSTRACT:

Big data in education promotes access to the analysis of learning behavior, yielding many valuable analysis results. However, with obscure and insufficient guidelines commonly followed when applying the analysis results, it is difficult to translate information knowledge into actionable strategies for educational practices. This study aimed to solve this problem by utilizing the learning analytics (LA) framework. We proposed a learning analytics framework based on human-centered Artificial Intelligence (AI) and emphasized its analysis result application step, highlighting the function of this step to transform the analysis results into the most suitable application strategy. To this end, we first integrated evidence-driven education for precise AI analytics and application, which is one of the core ideas of human-centered AI (HAI), into the framework design for its analysis result application step. In addition, a cognitive load test was included in the design. Second, to verify the effectiveness of the proposed framework and application strategy, two independent experiments were carried out, while machine learning and statistical data analysis tools were used to analyze the emerging data. Finally, the results of the first experiment revealed a learning strategy that best matched the analysis results through the application step in the framework. Further, we conclude that students who applied the learning strategy achieved better learning results in the second experiment. Specifically, the second experimental results also show that there was no burden on cognitive load for the students who applied the learning strategy, in comparison with those who did not.


Keywords:

Learning analytics framework, Analysis result application, Human-center AI, Learning strategy

Alwyn Vwen Yen Lee

National Institute of Education, Nanyang Technological University, Singapore // alwyn.lee@nie.edu.sg

Andrés Carlos Luco

Nanyang Technological University, Singapore // acluco@ntu.edu.sg

Seng Chee Tan

National Institute of Education, Nanyang Technological University, Singapore // sengchee.tan@nie.edu.sg


ABSTRACT:

Although artificial Intelligence (AI) is prevalent and impacts facets of daily life, there is limited research on responsible and humanistic design, implementation, and evaluation of AI, especially in the field of education. Afterall, learning is inherently a social endeavor involving human interactions, rendering the need for AI designs to be approached from a humanistic perspective, or human-centered AI (HAI). This study focuses on the use of essays as a principal means for assessing learning outcomes, through students’ writing in subjects that require arguments and justifications, such as ethics and moral reasoning. We considered AI with a human and student-centric design for formative assessment, using an automated essay scoring (AES) and feedback system to address issues of running an online course with large enrolment and to provide efficient feedback to students with substantial time savings for the instructor. The development of the AES system occurred over four phases as part of an iterative design cycle. A mixed-method approach was used, allowing instructors to qualitatively code subsets of data for training a machine learning model based on the Random Forest algorithm. This model was subsequently used to automatically score more essays at scale. Findings show substantial agreement on inter-rater reliability before model training was conducted with acceptable training accuracy. The AES system’s performance was slightly less accurate than human raters but is improvable over multiple iterations of the iterative design cycle. This system has allowed instructors to provide formative feedback, which was not possible in previous runs of the course.


Keywords:

Automated essay grading, Human-centric AI, Formative feedback, Machine learning, Ethics education

Pin-Hsuan Wang, Anna YuQing Huang, Yen-Hsun Huang, Ying-Ying Yang, Jiing-Feng Lirng, Tzu-Hao Li, Ming-Chih Hou, Chen-Huan Chen, Albert ChihChieh Yang, Chi-Hung Lin and Wayne Huey-Herng Sheu

Pin-Hsuan Wang

Department of Medical Education, Clinical Innovation Center, Medical Innovation and Research Office, Taipei Veterans General Hospital, Taipei, Taiwan // Taipei Veterans General Hospital, Taipei, Taiwan // karenwang0607@gmail.com

Anna YuQing Huang

Computer Science & Information Engineering, National Central University, Taoyuan City, Taiwan // anna.yuqing@gmail.com

Yen-Hsun Huang

College of Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan // michaelyhhuang@gmail.com

Ying-Ying Yang

Department of Medical Education, Clinical Innovation Center, Medical Innovation and Research Office, Taipei Veterans General Hospital, Taipei, Taiwan // College of Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan // yangyy@vghtpe.gov.tw

Jiing-Feng Lirng

College of Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan // jflirng@vghtpe.gov.tw

Tzu-Hao Li

College of Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan // Division of Allergy, Immunology, and Rheumatology, Department of Internal Medicine, Shin Kong Wu Ho-Su Memorial Hospital // pearharry@yahoo.com.tw

Ming-Chih Hou

Taipei Veterans General Hospital, Taipei, Taiwan // College of Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan // mchou@vghtpe.gov.tw

Chen-Huan Chen

Department of Medical Education, Clinical Innovation Center, Medical Innovation and Research Office, Taipei Veterans General Hospital, Taipei, Taiwan // College of Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan // chencc101@gmail.com

Albert ChihChieh Yang

Taipei Veterans General Hospital, Taipei, Taiwan // College of Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan // accyang@nycu.edu.tw

Chi-Hung Lin

College of Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan //linch.ym@gmail.com

Wayne Huey-Herng Sheu

Taipei Veterans General Hospital, Taipei, Taiwan // College of Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan // whhsheu@vghtpe.gov.tw


ABSTRACT:

Medical personnel need to learn occupational safety knowledge in clinical workplaces, not only to ensure their own safety, but also to further ensure patients safety. Based on Human-centered artificial intelligence (HAI) technology, this study will provide HAI-based occupational safety training system for two training topics, Needle Stick/Sharps Injury (NSSI) prevention and appropriate Clinical Waste Management (CWM). From April 2018 to December 2021, this clinical occupational safety HAI training is used by 342 medical personnel (doctors and non-doctors). This study aims to investigate the learning performance and effectiveness including decreasing anxiety and increasing mastering level of users. This study shows that, for the first-time and feel-friendly users of this HAI training system, not only can they achieve significant learning improvement, but they can also effectively decrease their anxiety and increase their mastery level of clinical work safety knowledge and skill. In terms of learning performance and effectiveness, this study found that doctors are significantly benefited by the HAI training system in contrast to non-doctors.


Keywords:

Clinical waste management, Needle stick sharp injury, Virtual reality

Xieling Chen

School of Information Technology in Education, South China Normal University, Guangzhou, China // xielingchen0708@gmail.com

Gary Cheng

Department of Mathematics and Information Technology, The Education University of Hong Kong, Hong Kong SAR // chengks@eduhk.hk

Di Zou

Department of English Language Education, The Education University of Hong Kong, Hong Kong SAR // dizoudaisy@gmail.com

Baichang Zhong

School of Information Technology in Education, South China Normal University, Guangzhou, China // zhongbc@163.com

Haoran Xie

Department of Computing and Decision Sciences, Lingnan University, Hong Kong SAR // hrxie2@gmail.com


ABSTRACT:

As a human-friendly system, the artificial intelligence (AI) robot is one of the critical applications in promoting precision education. Alongside the call for humanity-oriented applications in education, AI robot-supported precision education has developed into an active field, with increasing literature available. This study aimed to comprehensively analyze directions taken in the past in this research field to interpret a roadmap for future work. By adopting structural topic modeling, the Mann-Kendall trend test, and keyword analysis, we investigated the research topics and their dynamics in the field based on literature collected from Web of Science and Scopus databases up to 2021. Results showed that AI robots and chatbots had been widely used in different subject areas (e.g., early education, STEM education, medical, nursing, and healthcare education, and language education) for promoting collaborative learning, mobile/game-based learning, distance learning, and affective learning. However, a limited practice in developing true human-centered AI (HCAI)-supported educational robots is available. To advance HCAI in education and its application in educational robots for precision education, we suggested involving humans in AI robot design, thinking of individual learners, testing, and understanding the learner–AI robot interaction, taking an HCAI multidisciplinary approach in robot system development, and providing sufficient technical support for instructors during robot implementation.


Keywords:

Artificial intelligence robots, Topic modeling, Bibliometric analysis, Precision education, Research topics, Future of human-centered artificial intelligence

Shijin Li

Department of Education Information Technology, East China Normal University, China // shijinliEdu@163.com

Xiaoqing Gu

Department of Education Information Technology, East China Normal University, China // xqgu@ses.ecnu.edu.cn


ABSTRACT:

With artificial intelligence (AI) is extensively applied in education, human-centered AI (HCAI) has become an active field. There although has been increasing concern about how to systematically enhance the AI applications effect, AI risk governance in HCAI education has not been discussed yet. This study adopted literature meta-analysis, along with the Delphi and analytic hierarchy process (AHP) methods in order to establish the risk framework and calculate the index weight of HCAI education. The results confirm that the risk framework includes eight indicators, which respectively are misunderstanding of the HCAI concept (MC), misuse of AI resources (MR), mismatching of AI pedagogy (MP), privacy security risk (PSR), transparency risk (TR), accountability risk (AR), bias risk (BR), and perceived risk (PR). Meanwhile, the eight indicators are divided into four categories such as HCAI concept, application process, ethical security, and man-machine interaction. Moreover, the trend of risks types indicates that more than half of the articles consider only three or less risks types, and the evolution results of risks indicators gradually increased between 2010 and 2021. Additionally, the weights of the eight indicators are MP > MR > AR > PSR > TR > PR > BR > MC. Results obtained could provide theoretical evidence and development suggestions for future scientific governance of HCAI education. Furthermore, the risk framework not only systematically considers the risk governance order of HCAI education, but more importantly, it is the key bridge to the collaborative advancement of stakeholders such as managers, teachers, students, and parents, which can contribute to the scientific, healthy, and sustainable HCAI education.


Keywords:

Human-centered artificial intelligence (HCAI), Risk framework, Index weight, AHP, Delphi

Hui-Tzu Chang, Chia-Yu Lin, Wei-Bin Jheng, Shih-Hsu Chen, Hsien-Hua Wu, Fang-Ching Tseng and Li-Chun Wang

Hui-Tzu Chang

Center for Institutional Research and Data Analytics, National Yang Ming Chiao Tung University, Hsinchu // simple@nycu.edu.tw

Chia-Yu Lin

Department of Computer Science and Information Engineering, National Central University, Taoyuan // sallylin0121@gmail.com

Wei-Bin Jheng

Department of Computer Science, National Yang Ming Chiao Tung University, Hsinchu // sozai97.cs06@nctu.edu.tw

Shih-Hsu Chen

Department of Electrical and Computer Engineering, National Yang Ming Chiao Tung University, Hsinchu // brian880403@gmail.com

Hsien-Hua Wu

Department of Transportation and Logistics Management, National Yang Ming Chiao Tung University, Hsinchu // simonwu.mg09@nycu.edu.tw

Fang-Ching Tseng

Electrical Engineering and Computer Science Undergraduate Honors Program, National Yang Ming Chiao Tung University, Hsinchu // claire.eecs07@nctu.edu.tw

Li-Chun Wang

Department of Electrical and Computer Engineering, National Yang Ming Chiao Tung University, Hsinchu // wang@nycu.edu.tw


ABSTRACT:

The objective of this research is based on human-centered AI in education to develop a personalized hybrid course recommendation system (PHCRS) to assist students with course selection decisions from different departments. The system integrates three recommendation methods, item-based, user-based and content-based filtering, and then optimizes the weights of the parameters by using a genetic algorithm to enhance the prediction accuracy. First, we collect the course syllabi and tag each course from twelve departments for the academic years of 2015 to 2020. Next, we use the course tags, student course selection records and grades to train the recommendation model. To evaluate the prediction accuracy, we conduct an experiment on 1490 different courses selected by 5662 students from the twelve departments and then use the root-mean-squared error and the normalized discounted cumulative gain. The results show that the influence of item-based filtering on the course recommendation results is higher than that of user- and content-based filtering, and the genetic algorithm can find the optimal solution and the corresponding parameter settings. We also invite 61 undergraduate students to test our system, complete a questionnaire and provide their grades. Overall, 83.60% of students are more interested in courses at the top of the recommendation lists. The students are more autonomously motivated rather than holding extrinsic informational motivation across the hybrid recommendation method. Finally, we conclude that PHCRS can be applied to all students by tuning the optimal weights for each course selection factor for each department, providing the best course combinations for students’ reference.


Keywords:

Human-centered AI in education, AI course recommendation system, Learning aids in systems

Ting-Chia Hsu

Department of Technology Application and Human Resource Development, National Taiwan Normal University, Taiwan // ckhsu@ntnu.edu.tw

Hsiu-Ling Huang

Graduate Institute of Digital Learning and Education, National Taiwan University of Science and Technology, Taiwan // hsiuling427@gmail.com

Gwo-Jen Hwang

Graduate Institute of Digital Learning and Education, National Taiwan University of Science and Technology, Taiwan // gjhwang.academic@gmail.com

Mu-Sheng Chen

Department of Technology Application and Human Resource Development, National Taiwan Normal University, Taiwan // mushengchen946@gmail.com


ABSTRACT:

In traditional instruction, teachers generally deliver the content of textbooks to students via lectures, making teaching activities lack vibrancy. Moreover, in such a one-to-many teaching mode, the teacher is usually unable to check on individual students’ learning status or to provide immediate feedback to resolve their learning problems. Chatbots provide an opportunity to address this problem. However, conventional chatbots generally serve as information providers (i.e., providing relevant information by matching keywords in a conversation) rather than as decision-making advisors (i.e., using a knowledge-base with a decision-making mechanism to help users solve problems). Thus, this study proposes an expert decision-making-based chatbot to facilitate individual students’ construction of knowledge during the learning process. A quasi-experiment was conducted to compare the differences in the performances and perceptions of students using the expert decision-making-based chatbot (EDM-chatbot) and the conventional chatbot (C-chatbot) in the activities of a geography course. One class of 35 students was the experimental group, using the EDM-chatbot. The other class of 35 students was the control group, using the C-chatbot. The results of the study showed that the EDM-chatbot combined with expert decision-making knowledge significantly improved students’ learning achievement and learning enjoyment as well as reducing their learning anxiety, showing the value of the proposed approach.


Keywords:

Artificial Intelligence in Education, Expert knowledge, Decision tree, Chatbot, Interactive learning system

Tzu-Chi Yang

Institute of Education, National Yang Ming Chiao Tung University, Taiwan // tcyang.academic@gmail.com


ABSTRACT:

The development of digital competence has become an important part of higher education, and digital competence assessments have attracted considerable attention and concerns. Previous studies in this area mainly focused on self-reporting and manual review methods such as questionnaires, which offer limited assessment value. To solve this issue, this study uses natural language processing (NLP)—a current promising artificial intelligence (AI) technology—to analyze syllabi for assessing digital competence in universities. Analysis results show that the proposed method can achieve an average accuracy and consistency of over 80% with excellent efficiency. Moreover, the method demonstrates high consistency with manual evaluation results (kappa > 0.6) and enables automated large-scale objective assessment. In brief, the results suggest that the proposed method is efficient, effective, and reliable, making it a valuable solution for digital competence assessment. We accordingly explore the application expansion of this method in building the digital competence of universities. Furthermore, we discuss the theoretical, methodological, and applied contributions of this study.


Keywords:

Digital competence, Artificial intelligence, Higher education, Text classification, Machine learning

Starting from Volume 17 Issue 4, all published articles of the journal of Educational Technology & Society are available under Creative Commons CC-BY-ND-NC 3.0 license.