Building the Future of Education with Learning Analytics, Personalization and Collaboration An Interview with Dr Elvira Popescu

 


Bulletin of the Technical Committee on Learning Technology (ISSN: 2306-0212)
Volume 24, Number 1, 1-6 (2024)
Received Jan 19, 2024
Accepted Jan 21, 2024
Published online October 15, 2024
This work is under Creative Commons CC-BY-NC 3.0 license. For more information, see Creative Commons License


Authors:

Tamara Marie Barreto1email and Cathy Cheung2email

1: PCCE, Goa, India
2: Australian National University (ANU), Canberra, Australia

Abstract:

Embracing the future of education involves harnessing the power of learning analytics, personalization, and collaboration. Learning analytics enables educators to glean valuable insights from student data, tailors learning experiences to individual needs, and fosters interactive learning environments.

Dr. Elvira Popescu is a distinguished Full Professor at the Computers and Information Technology Department, University of Craiova, Romania. Her extensive research portfolio encompasses technology-enhanced learning, adaptive educational systems, learner modeling, computer-supported collaborative learning, learning analytics, and intelligent and distributed computing.

Keywords: Expert Interview, Personalized Education, Intelligent Learning Environment, Learning Style, Learning Analytics

I. INTRODUCTION

Education with Learning Analytics, Personalization, and Collaboration envisions a transformative approach to learning by integrating cutting-edge technologies. Learning analytics, involves the systematic analysis of student data to gain insights into learning behaviors, enabling educators to make informed decisions and interventions. Personalization, tailors the learning experience to individual students’ needs through adaptive technologies, ensuring a customized and effective educational journey. Lastly, collaboration emphasizes the importance of interactive and cooperative learning environments, facilitated by technology to encourage teamwork and communication skills. Together, these elements create a dynamic educational ecosystem that leverages data-driven insights, adapts to individual learning styles, and fosters collaborative skills, preparing students for the challenges of the future.

Dr Elvira Popescu has authored and co-authored over 100 publications, including two books, journal articles, book chapters, and conference papers. In addition, she co-edited six journal special issues, as well as 20 international conference proceedings. She participated in over 15 national and international research projects, three of which as a principal investigator. Prof. Popescu also serves as the Vice Chair for the IEEE Women in Engineering Romania Section Affinity Group and is a board member for the IEEE Technical Committee on Learning Technology and the International Association of Smart Learning Environments. She is also a Distinguished Speaker in IEEE Computer Society Distinguished Visitors Program (2020-2023), and she gave several invited talks at international conferences. She received several scholarships and awards, including five best paper distinctions. She is actively involved in the research community by serving as associate editor for three journals, member of five other journal editorial boards, organizing a series of international workshops (SPeL 2008–2020), serving as a conference chair, program committee chair and track chair for over 20 conferences.

II. Related Work

There are numerous research efforts exploring the development of models for student performance prediction, personalized learning experiences, and adaptive systems. Learning Analytics delves into the effective implementation of learning analytics in diverse educational settings. We hope this interview can inspire researchers in the field and provide future research directions.

III. The Interview

A. Personalized Education in the Modern World

Q: What are the benefits of implementing personalized education?

A: Most educational institutions tend to follow a teacher-centred, subject-specific approach to teaching. The textbook matter is solely taught to a large group of students at a time. In stark contrast, personalized instruction accommodates each student’s learning style. Through her intensive research on learning-style based education, Dr. Popescu discusses how implementing personalized education may be advantageous:

“Most educational institutions nowadays use the so-called Learning Management Systems, such as Moodle for example, which is one of the most popular. Actually these are integrated systems which are very useful. They offer support for a wide range of activities in the educational process. But the problem is that they don’t provide personalized services, so all students are given access to the same set of educational resources, the same tools, without taking into account the differences between the students in terms of knowledge level or background, motivation, interests and so on. So adaptive educational systems try to offer an alternative to this non-individualized approach by providing some services which are adapted to the learner profile.

One of the most important characteristics in the learner profile, as you have mentioned, is the learning style of the student – that is, the individual preferences of the student towards learning. Identifying and understanding the learning style of the student can have a positive impact on learning. So on one hand, it could for example increase the self- awareness of the students. So, this would allow them to take advantage of their strengths and to minimize the effect of their weaknesses by means of self-regulation.

On the other hand, the teachers can also use the knowledge of their student’s learning styles to provide activities to reach those students whose needs are not met by the standard curriculum. And of course, adaptive educational systems can also offer personalized learning experiences based on the student’s learning style and this adaptation can bring various benefits, as discussed. So we can have an increase in the student’s satisfaction and engagement, we can have higher learning gain and a decrease in the time needed to learn. So all of these definitely mean an improvement in the educational experience.”

Q: How does the intelligent learning environment, WELSA, help meet the varied needs of the students?

A: With so many people trying to learn more, each having unique learning styles and preferences, it is quite challenging to tailor the teaching methods to fit each one’s needs. Thus, Dr. Popescu’s research proposes WELSA, an intelligent learning environment that adapts to the learning style of the students:

“This system is based on a so-called ‘Unified Learning Style Model’ that we proposed, which integrates characteristics from several models found in the literature related to perception modality, processing and organizing information, social aspects, motivational aspects, and so on. This model has several advantages: on one hand it solves the problem related to the fact that we have so many learning style models and the fact that there are overlapping concepts between them. On the other hand, it provides a feature-based modelling approach – so this is simpler and more accurate than the stereotype-based, traditional modelling approach. And also this offers the possibility to have more effective, finer grained adaptation actions.

So, what did we do? First of all, we proposed an implicit learner modelling method based only on the interpretation of the student’s actions. So, we provided an automated approach in which the characteristics from the Unified Learning Style Model were identified by monitoring and analysing the learner’s behaviour in the educational system. This means that we did not require any additional effort from the part of the students. So they didn’t have to fill in any kind of learning style questionnaire which, by the way, might have come also with some reliability and validity problems of its own.

Secondly, we identified the adaptation technologies that were best suited for students with different learning preferences and we defined the corresponding adaptation rules. So WELSA performs a dynamic adaptation, by generating the individualized web pages for each student in an automatic way. Thus the system is able to include a large number of learning preferences, but without a large increase in the teacher’s workload, to answer your specific concern. So the teacher will have to prepare the same amount of educational materials which will be dynamically combined by the system according to each student’s preference.

And of course, we tried to use this system in practice and we obtained quite good results. Most of the students evaluated their learning experience with WELSA as very positive. The adaptation managed to improve the learner’s enjoyment, overall satisfaction, motivation and learning effort.”

Q: Many educational institutions are still adopting traditional approaches today. How would you encourage them to experiment with new approaches and systems?

A: It is clear that accommodating learning styles in the learning process is beneficial. However, most educational institutions have not changed their approach. In fact, many are skeptical about shifting away from the traditional methods:

“Yes, indeed. Unfortunately, many of the adaptive educational solutions that were proposed so far did not move past the stage of research prototypes, so they are used on a small scale only, in the department or the institution of the researchers who designed them. But I think that this is slowly changing since the educational landscape evolves, so learning gets more structured and trying to accommodate the needs of more diverse student audiences. So, in this context, trying to apply personalized learning by adapting the educational experience to the learner’s preferences and needs does become critical. There is indeed a challenge and this consists in providing an alternative to the one-size-fits-all approach in designing the learning technologies and taking into consideration the individual differences among the learners and the various contexts in which the learning takes place.

And there are indeed many questions that need to be addressed, such as: What can we adapt to? What can be adapted? How can we collect and process the relevant data? What is the impact of adaptation? How can we promote personalized learning in real contexts? How can we support the educators to develop personalized content and resources? How do the learners perceive personalization? or How can the communities of learners benefit from this personalization? Actually a forum in which we are trying to find answers to these questions is the track on “Adaptive and Personalised Technology-Enhanced Learning” that we are organizing every year at the ICALT conference (IEEE International Conference on Advanced Learning Technologies), which is actually supported by the IEEE Technical Community on Learning Technology (TCLT).”

Q: Will shifting online compromise all the social benefits that are essential to one’s growth and learning?

A: Humans are inherently social. As we grow up, most of us spend our years learning together with our classmates. This promotes discussion, creativity and collaboration. Thus, Dr. Popescu elaborates on how we can obtain these social benefits in the online mode:

“Well, not necessarily. But as you pointed out, learning is at least as much about access to other people as it is about access to information. So access to peers, access to other learners is extremely important. So this basically lies at the heart of the social learning environments. And these kinds of social learning environments aim to provide the learners with a medium in which they can engage with each other, they can engage with the teachers, they can share experiences, co-create knowledge and of course, work and learn collaboratively.

Actually, we should not forget the fact that the generation of students that we are teaching today, which is actually your generation, was raised in the context of digital technologies, so in a world of communication, of wide availability of information. And these so-called ‘digital natives’ are said to have different patterns of work, attention and different learning preferences. So that’s why the traditional teaching methods should be adapted as much as possible to this new ‘Internet generation’ and offering support for social learning is one of these adaptation mechanisms.

And one of the ways to do this is to introduce social media tools in education. So social learning environments can rely on this kind of social media tools or Web 2.0 technologies. This is because their underlying principles (such as user-centred, participative architecture, openness, interaction, social networks and collaboration) are in line with modern educational theories such as socio-constructivism. So, according to this, knowledge cannot be simply transmitted, but it has to be constructed by the individual, many times by means of collaborative efforts in groups of learners.

And also in this context, by using social media, the user is not just content consumer, but also content generator, oftentimes in a collaborative manner. And again, this is in line with some modern contribution-based pedagogies which state that creating learning resources in a collaborative way and sharing them with others is an important and promising practice for learning efficiently.

Indeed, in recent years and also during the past couple of years due to the pandemic, the social media tools have found their way into the educational landscape and there were some encouraging results in terms of student satisfaction, knowledge gain or learning efficiency. And actually, there have also been proposed some special purpose social media tools that were built specifically for educational use.
These technologies can be used to foster the communication and collaboration between learners and help them to learn collaboratively, to create online learning networks.”

Q: Could you tell us more about your proposition of a social learning environment?

A: Dr. Popescu proposed a remarkable social learning environment that provides valuable services to the teachers as well as students:

“Yes. So we proposed such a social learning environment, which . we called eMUSE – it comes from ‘empowering MashUps for Social E-learning’. So basically the name comes from the technology underlying the system – the social media tools were integrated in the platform by means of mash-ups.

So what did we do? We included several social media tools that the teacher could select from and decide which one they wanted to use in their course. So we included a blog, a wiki, a social bookmarking tool, a microblogging tool and several social media tools. And the eMUSE platform provides an integrated access to the social media tools which are selected by the instructor. So it provides a common access point, detailed usage instructions, a summary of the latest activity, some basic administrative services and also some functionalities for evaluation and grading. And the most important thing, the platform retrieves the students’ actions with each tool and stores them in a local database and then it offers a summary of each student’s activity, various graphical visualizations, comparisons with peers, evolution over time and so on.

And we used this platform in a project-based learning scenario. Since you may not be familiar with it, project-based learning is a student-centred instructional approach in which learning is organized around projects. And these projects have to be complex and include challenging and authentic tasks in which the students have to work relatively autonomously, so the teacher only must play the role of facilitator. The students basically have to collaborate in various activities in problem solving and decision making with the final goal to design a realistic product or presentation.

So in our case, we applied it in conjunction with our social learning environment, eMUSE. And we found out that the results were pretty encouraging. The majority of the students were satisfied with using this platform and they were willing and eager to repeat this kind of experience for other courses in the future.”

B. Learning Analytics Techniques

Q: Which of the algorithms and techniques used in Learning Analytics are the most effective and why?

A: Learning Analytics aims at analysing student data to find patterns in student’s behaviour to predict their performance and thus optimize the educational platform. Dr. Popescu’s work suggests that analysis techniques such as classification, regression, clustering and PCA algorithms can be used to do so:

“Yes, first let me tell you a bit more about the context of our research. It is actually related to the social learning environment that I have just mentioned, eMUSE. So we used the platform for several years, for several student cohorts at the University of Craiova in Romania. And our study took place in the context of an undergraduate course for Computer Science students – it was a Web Applications Design course. And the instructional approach was project-based learning, as I mentioned. So the students had to collaborate in teams of around four peers to build a complex web application of their choice. Some decided to implement a virtual bookstore, others a professional social network or an online travel agency and so on.

We used a blended learning approach, so we had face-to-face meetings each week between each team and the instructor for checking the progress of the project and for providing some feedback, for answering some questions. For the rest of the time, students had to rely on the social media tools integrated in eMUSE as support for communication and collaboration activities. So basically MediaWiki was used as a tool for collaborative writing tasks and for organizing the team knowledge and resources. Blogger was used for reporting the progress of the project, so like a kind of learning diary, but also to provide some feedback and some solutions for the peer problems. Each team had its own blog, but we also encouraged the collaboration between the teams. And then we also have Twitter, which was meant for shorter communication, for some additional connections between the peers, also for posting some short announcements, some news regarding each project.

And all the student data, so all their contributions on the social media tools, were collected by eMUSE. And the point was to analyse, of course, this data. One of the goals was to predict the student’s performance based on the behaviour on the social media tools.

So, we computed a set of numeric features for each student which cover relevant characteristics. These included, for example, the number of posts and edits, the length of the content, the time distribution of the messages and so on. And our aim was to predict the final grade based on this set of features. So we applied various algorithms, as you mentioned, various classification and regression algorithms, including KNN, neural networks, decision trees, support vector machines. But we also applied a novel algorithm, called ‘Large Margin Nearest Neighbour Regression’, which was proposed by a colleague of mine, Professor Florin Leon from Gheorghe Asachi Technical University of Iasi, also in Romania.

And the results were very good. So our algorithm managed to outperform the other algorithms on all the datasets. And we obtained also quite good correlation coefficients and more than 85% of the predictions were within one point of the actual grade. Thus, we consider these to be very good results. So basically our study showed that the student’s actions on the social media tools are good predictors of academic performance. Which, from a pedagogical point of view, means that, as a general rule, a higher engagement with social media tools correlates with a higher final grade. This makes sense and it is in line with several other previous studies which also found that online participation is a strong indicator of student performance and can improve the learning effectiveness.”

Q: How does Educational Data Mining (EDM) tap into the power of Big Data in the era of online education?

A: When the pandemic struck, the world switched to online-learning. From lectures and meetings to Massive Open Online Courses (MOOC), more people have started learning online than ever before. Every learner generates an immense amount of data, which can be used to gain a deeper understanding of the nature of learning and teaching online.

From her research, Dr Popescu provides some insights on how to utilise this data:

“Well, it’s difficult to give a general answer, but I can describe some of the methods that we applied in the context of the social learning environment, because this was the context of our research. So, as I was saying, we gathered quite a lot of data with eMUSE and our objective was to analyse this data from multiple perspectives. So apart from the academic performance prediction that I have discussed, we also explored other things such as the relationships between the students’ learning styles and their social media use.

Also, we investigated the students’ collaboration patterns and also the community of inquiry supported by social media tools. So more specifically, we tackled four main research directions: analytics dashboards, predictive analytics, social network analytics, and discourse analytics.

And with respect to the techniques used, we applied various approaches, as I just mentioned: classification, regression, clustering, principal component analysis (PCA) algorithms, but also textual complexity analysis, social network analysis techniques and content analysis based on the Community of Inquiry model. So overall, I can say that we addressed the so-called ‘trinity of methodological approaches’: network analysis (meaning the social relations), process-oriented analysis (based on action logs and pattern detection), and content-oriented analysis (based on the actual content created by the students). And of course each of them has their specific role and they all helped us to get a more in-depth and comprehensive understanding of the learning process.”

C. The Peer Assessment Approach

Q: What is the “Peer Assessment” approach for collaborative learning? What are its benefits in education?

A: With the rising popularity of collaborative learning, many institutions have begun adopting the ‘peer assessment’ approach. Dr. Popescu’s research provides a thorough analysis of what it is and why it is considered to be beneficial to the students as well as teachers:

“Peer assessment is indeed gaining popularity in recent years, especially in the context of collaborative learning.
What is it? Well, first of all, it is also known as ‘peer review’ or ‘peer evaluation’. It refers to the involvement of the students in the process of assessing the work of their fellow learners. So basically the students provide feedback and sometimes grades to their peers – and in this case the term ‘peer grading’ can be used.

So peer review has indeed many benefits, both for the provider and the receiver of the assessment. Those students who play the role of evaluators are exposed to the work and ideas of their peers and this can offer them some interesting new perspectives on the topic. It can help them extend their knowledge and their understanding. And by performing the evaluation, this can contribute to the development of critical thinking skills, reflection skills, and meta-cognitive skills in general. It also improves the motivation, responsibility and even the self-confidence of the evaluators.

On the other hand, students who receive the feedback from their peers also benefit. They get more detailed and timely feedback, as compared to the limited feedback which is provided by the instructor, especially if we talk about large classes. Actually, this refers also to the main benefit for the teachers: this means a decrease in the time that they need to allocate for evaluating the students’ activities.

But of course, there are also some potential pitfalls for peer assessment. There can be issues with the validity, reliability and fairness of the process, especially when we talk about peer grading. Also, some students may resent or dislike the evaluation of their peers’ work. They may find it too time-consuming, or they may lack confidence in their evaluation ability. Some students might not take the process too seriously unless it somehow counts for their final grade or if it is monitored by the instructor.

But on the whole, studies show that the engagement of the students is generally increased by means of peer evaluation. This is because their motivation for learning has a strong social dimension, so they are somehow inclined to pay more attention to the opinions and the feedback of their peers. And also there is an increased level of interactivity between the students and they tend to play a more active role in their learning.”

Q: Most existing approaches to peer assessment are rather context-specific. Could you share with us your research on a more generic peer assessment platform?

A: Over the past decades, many peer assessment platforms have surfaced. However, these are rather context-specific systems that are designed to support only a particular type of learning activity or discipline. In Dr. Popescu’s work, a more generic peer assessment platform was proposed:

“Yes. So together with one of my PhD students, Gabriel Badea from the University of Craiova, we designed and implemented, as you said, a more comprehensive and generic peer evaluation system, called LearnEval.
Of course, as you said, there are already many peer assessment platforms proposed in the literature. So first of all, we performed a thorough literature analysis. We found out that many of these systems were designed to support only a specific type of learning activity (for example, they were dedicated to writing assessments) or a specific discipline (for example, they were dedicated for learning programming languages) or a specific type of content (for example, they were dedicated to assessing online videos) and so on. And also, most of these platforms are focused on a limited set of functionalities, which makes sense as this was in accordance to their specific goal.

So our conclusion was that there was a need for a more generic platform which should integrate some of the useful functionalities which are already present in various existing systems, but at the same time add some novel desirable features. So we proposed LearnEval platform. It was meant to be a solution to the existing challenges in the current peer assessment systems. And we also aimed to fulfil the requirements of any type of course.

So the first step in developing the platform was to devise a peer assessment workflow that would be suitable in a wide range of pedagogical scenarios and contexts, meaning that it would be easy to be configured, it would be highly configurable according to the teacher’s needs. So the teacher could define various aspects related to the peer assessment process such as deadlines, assessment criteria, number of reviews per submission, the mechanism used for allocating the submissions to the reviewers, anonymity of reviewers and reviewees, the weights of the metrics that are used for computing the various scores of the learners. And the instructors can track the progress of the learners in order to detect some students who are at risk. They can also visualize the learner models and they can create calibration assignments in order to enhance the assessment skills of the students at the beginning of the course.

And of course, the platform also provides functionalities and a wide range of modules for the students: an Assignment module (where the student can view the information about the assignments and submit solutions), a Calibration module (where the student can practice and test and improve their assessment skills before starting to actually assess their peers’ work), a Review Solutions module, a Statistics module (where the student can explore various aspects related to their peer assessment activity) and also a Scores module (where they can visualize their personal scores and their open learner model).

And we extended and improved the platform throughout the years. The most important extension is an innovative dynamic review allocation mechanism which increases the fairness of the process and also the number of evaluations per submission. And also we proposed a hybrid approach for trying to discourage the so-called ‘rogue review’ behaviour which appeared for some of the learners.

And again, we have applied the platform in several courses, in several learning scenarios. We also made a comparative study between a mandatory and an optional peer assessment activity and we obtained good results. So, we got a good peer grading validity, which means relatively high correlation between the grades assigned by the platform based on the peer reviews and the grade given by the instructor. And also, generally, the students’ experience of using LearnEval was good in terms of usability and learner satisfaction.

D. Challenges in Implementing Technology-Enhanced Learning

Q: After having this insightful interview with you, Professor, we have gained a better understanding of the potential of technology-enhanced learning. To conclude, I would like to know what challenges we may face while adopting this approach and how we can overcome them.

A: “Yes, this is a tough question. So if we look through the various roadmaps, the various trends regarding educational technologies, we see that the most popular technologies, which are expected to have a big impact in the following years, include adaptive and personalized learning, remote and hybrid learning tools and spaces, open educational resources, microcredentialing, learning analytics, big data and in general the use of Artificial Intelligence (AI) for education. And there are various challenges ahead, not only from a technological point of view, but also from a social, economic, environmental and political perspective.

And maybe I should just focus a bit on the challenges that are related to learning analytics because we discussed this topic more before. So in this context, I would say that one such challenge is transferability. Since the students currently use various systems for learning (in different years of study, or even in different courses – some use Moodle, some use Google Classroom, some use other platforms and so on), it should be possible to transfer the student model from one system to the other and to use this information to improve the current student model. And of course, ideally, the data could be gathered not just from the educational environment, but also from some other sources, from the personal environment of the students, from different points of view, such as familiar, economical, emotional, psychological and so on.

And in the same vein, the learning analytics models should be generalizable, so they should be able to work across courses and across systems, ideally without having to refit them or to modify them. And of course they should also work on different populations, not just the convenience sample that we have at hand (generally these are students from our own university) – but they should work on completely different samples, with variations in race, ethnicity, citizenship, economic status, language and so on.

Also, models should be interpretable and scrutable, so predictions should be intuitive for the user, so that both the teachers and the learners would be able to understand them and trust them.

And in this context, of course, ethical and privacy issues are very important and also focusing on the student more – so instead of having a traditional institution-centric view, we should take more into account the perspectives of the students because actually they are the primary stakeholders. So students should be consulted in the development and the application of learning analytics and we should not just assume that we know what students want and what their concerns are, or how they would like the data to be presented.

And finally, we should not forget the emerging educational environments, which are based on innovative technologies, like mobile, ubiquitous, moving on to virtual reality, augmented reality environments and now the metaverse. So learning analytics should be applied in these types of cutting edge environments as well, which of course will come with their own set of challenges.”

IV. Conclusion and Reflection

In reflection, the comprehensive interview with Dr. Elvira Popescu sheds light on the transformative impact of personalized education and technology-enhanced learning. The discussion emphasizes the significance of adapting educational approaches to cater to individual learning styles, with a focus on Dr. Popescu’s work in the field of adaptive educational systems.

The implementation of intelligent learning environments, such as WELSA, and the proposed social learning environment, eMUSE, exemplifies innovative solutions addressing the diverse needs of students.

The interview further delves into the challenges and advancements in learning analytics, highlighting the need for transferability, generalizability, interpretability, and ethical considerations. Dr. Popescu emphasizes the importance of consulting students in the development and application of learning analytics, emphasizing a learner-centric perspective. The integration of adaptive technologies and peer assessment platforms, like LearnEval, showcases practical implementations of research findings.

Addressing the concerns surrounding online learning, Dr. Popescu emphasizes the importance of maintaining social benefits in digital education through social learning environments. The interview concludes by outlining the challenges and future trends in educational technologies, emphasizing the need for generalizability, interpretability, ethical considerations, and adaptability in the evolving landscape, particularly in the context of emerging technologies like virtual reality, augmented reality, and the metaverse. Overall, the interview provides a comprehensive overview of the potential and challenges in leveraging technology to enhance the educational experience.

 

Reference

[1] Popescu, “Adaptation provisioning with respect to learning styles in a Web‐based educational system: an experimental study,” Journal of Computer Assisted Learning, vol. 26, no. 4, pp. 243–257, 2010.

[2] Bernard, T. W. Chang, E. Popescu, and S. Graf, “Using artificial neural networks to identify learning styles,” in Artificial Intelligence in Education: 17th International Conference (AIED 2015), Madrid, Spain, 2015, pp. 541–544, 10.1007/978-3-319-19773-9_57.

[3] Popescu and D. Cioiu, “eMUSE – Integrating Web 2.0 Tools in a Social Learning Environment,” in Lecture Notes in Computer Science, vol. 7048, pp. 41-50, 2011, 10.1007/978-3-642-25813-8_5.

[4] Popescu, C. Bädicä and L. Moraret, “WELSA: An Intelligent and Adaptive Web-Based Educational System,” in Intelligent Distributed Computing III. Studies in Computational Intelligence, Berlin, GER:Springer, vol. 237, pp. 175-185, 2009.

[5] Popescu, “Providing collaborative learning support with social media in an integrated environment”, in World Wide Web, vol. 17, no. 2, pp. 199-212, Mar. 2014.

[6] Bernard, T. W. Chang, E. Popescu, and S. Graf, “Learning style Identifier: Improving the precision of learning style identification through computational intelligence algorithms,” in Expert Systems with Applications, vol. 75, pp. 94-108.

[7] Popescu, “Diagnosing students’ learning style in an educational hypermedia system,” in Cognitive and Emotional Processes in Web-based Education: Integrating Human Factors and Personalization, IGI Global, pp. 187-208.

[8] Popescu, C. Bădică, and L. Moraret, “Accommodating learning styles in an adaptive educational system,” in Informatica, vol. 34, no. 4, pp. 451-462.


All authors contributed equally to this work.

The interview recording can be found on the IEEE TCLT YouTube Channel (https://youtu.be/huPRC0WHXkM?si=IXjvqVBv9zudGoT2).


 Authors

Tamara Marie Barreto

Tamara Marie Barreto

holds a Bachelor’s Degree with Honors in Computer Engineering from PCCE, Goa, India. Her interests include Data Science and Analytics, Machine Learning and Blockchain. She is actively involved in hands-on projects and consistently enhances her skills and knowledge.

Cathy Cheung

Cathy Cheung

a recent graduate with a Master of Computing from the Australian National University (ANU), has transitioned from a background in real estate to computing driven by her passion for the field. She is currently working as a software developer and a casual computing tutor at ANU.