In higher education, to cultivate exceptional individuals equipped with the competencies to navigate an unpredictable society, continuous quality assurance and enhancement of education in higher education institutions are required. To satisfy these requirements, it is essential to implement institution-wide, top-down initiatives through teaching and learning management, rather than relying solely on individual efforts such as course-level improvements. However, within such institution-wide initiatives, improving teaching and supporting student learning through faculty-development initiatives is also essential, although sustaining these initiatives remains challenging.

In the context of quality assurance and enhancement of education in higher education institutions, it is important to identify what students have learned and what competencies they have developed. “E-portfolios”—which include learning artifacts such as reports, creative works, photos and videos capturing learning processes, and learning journals—are gaining considerable attention as a tool for visualizing these learning outcomes. If learners register their e-portfolios in the e-portfolio system, and the system can promote reflection on learning while analyzing and visualizing learning outcomes, learners will be able to understand more fully what they have learned and the competencies they have developed. In this way, as learners actively engage in their own learning, quality assurance and enhancement of education in higher education institutions can be achieved through a bottom-up approach that places the learner at the center. In this study, we developed an e-portfolio system for analyzing and visualizing competency trends from learners’ artifacts and learning journals as evidence of learning outcomes.

The system consists of three functions.
- Promoting reflection on learning related to the e-portfolio:
A task to describe the e-portfolio registered is assigned to a generative AI, and the AI revises example reflection prompts. The AI then generates reflection prompts.
- Visualizing learning progress related to the e-portfolio:
Based on the Standards for the Establishment of Universities in Japan, a task to classify what the learner has studied from the registered e-portfolios and their associated learning journals is assigned to the generative AI, and the learning progress is visualized.
- Determining and visualizing the competencies developed through the e-portfolio:
A supervised learning model is used to extract descriptions related to the development of competencies. Morphological analysis is then used to determine the competency to which each extracted description corresponds to. The number of extracted descriptions can also be visualized.

The system operates as follows. At the end of a learning session or other appropriate time for reflection, learners are prompted to register their e-portfolios created during the learning process. The generative AI is instructed to read and describe the learner’s portfolio. Prompts are then generated and presented. The learner reflects on what they have become able to do and how they have changed. The generative AI then reads the e-portfolio and what has been learned is visualized. The competencies fostered through the learning journal are also determined and visualized. Finally, the learner reviews the visualized results and writes how to apply their learning in the future.

In the future, we plan to implement and evaluate the system.

Keywords: Quality assurance and enhancement of education, learning outcomes, teaching and learning management, e-portfolio, generative AI.