News and events archive

Professor Atsushi Shimada from Kyushu University, Japan, has visited the Faculty of Informatics, Masaryk University, to share his expertise at the intersection of learning analytics, machine learning, and multimedia processing. A leading researcher in educational data analysis and AI-driven learning environments, Professor Shimada has been recognized with numerous awards for his contributions to intelligent systems and human-centered computing. 

During his stay in Brno, he delivered a guest lecture titled “AI-Powered Learning Analytics for Transforming Education,” exploring how artificial intelligence can help personalize learning and support educators through data-driven insights. On this occasion, we spoke with Professor Shimada about his research, recent projects, and his perspective on the future of AI in education.

What motivated you to explore the integration of AI into learning analytics?

My initial motivation came from my background in image processing and pattern recognition, where I became fascinated by how data-driven approaches can reveal hidden structures. When I moved into the field of education, I realized that learning environments also generate a tremendous amount of data, and that analyzing these traces could provide valuable insights into how students learn and how we can better support them.

In your presentation, you share “an approach that combines predictive modeling, personalized feedback, and real-time monitoring with advanced media processing techniques for analyzing multimodal data such as text, video, and handwritten notes.“ Can you explain or simplify this for us to understand the process and what the outcome is for individual learners?

We use information from what students write, their notes, and their online activities to understand how they are learning. This helps us spot when someone might be struggling, so we can give them timely and personal feedback. For learners, the result is that they receive support just when they need it, making it easier to stay on track and motivated.

Could you share a concrete example where AI-powered learning analytics has already made a difference in a real educational setting?  

For example, in one of our studies we used AI to automatically generate summaries of learning materials. This helped students prepare more efficiently before class and improved their understanding of the content in advance. We also applied AI to provide real-time feedback during lessons, which allowed instructors to adjust their teaching on the spot. Together, these applications made learning more engaging and supportive for both students and instructors.

What are the biggest challenges you see when implementing AI-based tools in classrooms or online courses?

One of the biggest challenges is making sure that AI systems are reliable and transparent. Teachers and students need to trust the results and understand why a recommendation is made. Another challenge is fairness: we must ensure that these tools support all learners equally and do not unintentionally disadvantage some groups. Finally, there is the practical side: if AI systems are too complicated or add extra work, they will not be used effectively.

How do you balance the opportunities of personalization through AI with the risks around data privacy and student trust?

Personalization through AI can create powerful opportunities for supporting each learner in unique ways, but it only works if students trust the system. To balance this, we make sure to obtain students’ consent, use only the data that is truly needed, and keep it secure and anonymous. Just as importantly, we explain clearly how the data is used so that students feel in control rather than monitored. 

Japan is known for embracing new technologies in everyday life. Does this openness also extend to education, or do you encounter barriers when introducing AI into classrooms there?

It is true that Japan often embraces new technologies in daily life, but in education the picture is more mixed. When introducing AI into classrooms, there are barriers such as concerns about data privacy, strict curriculum requirements, and the additional workload for teachers. At the same time, we are seeing progress as digital textbooks, online learning platforms, and pilot projects using AI are becoming more common. While adoption in education tends to be more cautious, the openness is certainly growing.

In your experience, what role does the teacher still play when AI systems are introduced to support learning? 

Even when AI systems are introduced, the role of the teacher remains central. AI can provide data, feedback, and suggestions, but it is the teacher who interprets this information, understands the human side of learning, and decides how best to guide each student. In this way, AI serves to reinforce rather than replace the teacher’s role.

Over the next decade, what role do you expect AI to play in higher education worldwide, and what skills should today’s students and researchers develop to prepare for it?

Over the next decade, I expect AI to become a trusted partner in higher education, supporting personalized learning for students and helping educators design more effective courses. It will surely become an indispensable tool for advancing education, but it is equally important not to rely on it too heavily or lose sight of the true purpose of learning. To prepare for this future, today’s students and researchers should build basic AI literacy, think critically about the results, use data responsibly, and cultivate the human skills of collaboration and creativity that AI cannot replace.

Photo: Mahiro Ozaki, Atsushi Shimada, and Valdemar Švábenský

At FI MU, you are in touch with Dr. Valdemar Švábenský, who had spent over a year in your laboratory at a post-doctoral position. What research did you carry out together and do you foresee any future collaborations?

We worked together on learning analytics, focusing on how to represent learning activity data effectively and predict student behaviors. I look forward to continuing this collaboration in future international projects.

After experiencing FI MU and meeting our students and colleagues, what impression will you take back with you to Japan?

After experiencing FI MU and meeting many enthusiastic students and colleagues, I was deeply impressed by the open-minded atmosphere and the strong passion for advancing research and education through collaboration. Our discussions in Brno were particularly meaningful, as we exchanged ideas on learning analytics and educational data utilization. I was also very pleased to see promising possibilities for collaboration with FI MU’s strong initiatives in cybersecurity education. 

These conversations inspired me to envision new directions for connecting data-driven learning research with practical, secure, and human-centered educational systems. I strongly hope that this visit will further strengthen the collaboration between Masaryk University and Kyushu University, leading to deeper research exchange and joint initiatives in the near future.

Thank you for the interview.

—

Remarks from Valdemar Švábenský:

I was delighted that Prof. Shimada and Mr. Ozaki were able to visit Brno despite the long distance from Japan. Having previously worked with them at Kyushu University as a postdoctoral researcher, I truly appreciate the opportunity to continue our collaboration after returning to the Czech Republic. During their visit, we discussed plans for future joint research, including meeting at a flagship conference in 2026 and the possibility of a short research stay at Kyushu University. Although Prof. Shimada’s lecture took place in the middle of a busy semester, it attracted a large audience and sparked a lively discussion. Members of another research group at FI MU also discussed potential collaboration with Prof. Shimada’s laboratory, which I was pleased to see. I look forward to further strengthening our partnership with Kyushu University.

—

Atsushi Shimada received his M.E. degree in Information Science and Electrical Engineering from Kyushu University in 2004 and his D.E. degree from the same university in 2007. He is currently a Professor at the Faculty of Information Science and Electrical Engineering, Kyushu University. From 2015 to 2019, he served as a researcher for the JST PRESTO program. His current research interests include learning analytics, pattern recognition, media processing, and image processing. He has received numerous awards, including the MIRU Interactive Presentation Award (2011, 2017), MIRU Demonstration Award (2015), First Place in the Background Models Challenge 2012, the PRMU Research Award (2013), First Place in the SBM-RGBD Challenge (2017), the ITS Symposium Best Poster Award (2018), the JST PRESTO Interest Poster Award (2019), the IPSJ/IEEE Computer Society Young Computer Researcher Award (2019), the CELDA Best Paper Award (2019), and the MEXT Young Scientist Award (2020).