This paper explores the appropriate integration and utilization of generative AI within various educational and learning contexts. Regarding the research stage prior to the practical stage, many researchers in our society are trying to integrate AI technology into their system for enhancing the effectiveness for each educational purpose. In addition, when employing generative AI products in such educational settings with higher educational institutions, it is crucial to meticulously pay attention to licenses, terms of use, costs, and all aspects related to ethical and compliance requirements, ensuring awareness of the latest information. Presently, generative AI products and services are newly emerging, evolving, and expanding with increasingly complex and specific conditions. While these technologies have significant potential to generate compelling outputs in formats such as images, videos, and audio, this paper focuses on text-based products or services, which are the most widely applicable in educational technology researchers, focusing on human-computer interactions perspectives.

In online classes, it is important to facilitate learning so that students can engage in learning proactively by promoting them to understand learning contents and to grasp their learning status. One possible method for this is to provide formative feedback according to learning status. However, it is not easy to provide formative feedback by identifying gaps in students’ learning status in relation to lesson goals or evaluation criteria. The purpose of this study was to facilitate student learning in online classes. We focused on free-form assignments and developed a system that uses supervised learning to classify reports by learning status and provides formative feedback based on classification results. Evaluation results showed that the system could grasp the level of free-form assignments submitted by students to some degree. And it was suggested that providing formative feedback could facilitate students to reconsider how they approach assignments and engage in learning proactively.

In this paper, we explore the development and effectiveness of a system that leverages LLM (GPT) to generate personalized learning advice based on individual learners’ progress and reflections accumulated online. By constructing prompts that incorporate both quantitative and qualitative data, GPT generated the advice. The appropriateness of the generated advice was evaluated from both “teacher” and “learner” perspectives. The results indicated that the advice closely matched what a teacher might offer. Additionally, implementing the advising system in actual classes and evaluating it through surveys showed that learners generally set their goals for the next week based on the system’s advice with a sense of satisfaction. Thus, it was found that the generated advice was generally appropriate from both “teacher” and “learner” perspectives.

It is expected that learning at museum will cultivate sensitivity and critical thinking habits of learners. The present study developed a Prior learning by Faceted Search (PFS) model for museum learning that fosters awareness of hypothetical reasoning as critical thinking. In this study, a pre-visit learning program for history museum constructed based on the PFS model was implemented with university students. This study examined whether the program improves readiness for learning at museum visited after the program. Qualitative data analysis methods were used to analyze the reflective essays submitted in the educational practice. The results show that learning with the faceted search application can be helpful for prior learning to enhance the motivation for learning at the museum, along with the inspiration to learn the hypothetical reasoning process. These results indicate that the PFS model has the potential to contribute to the construction of effective pre-visit learning programs.