Data science education often lacks the kinds of experiences it truly requires.
Observing phenomena, interpreting data, analyzing it, generating ideas,
testing those ideas, and examining the results—
this full sequence of experience has become difficult to obtain in real environments.
Opportunities to observe are limited,
the data available is often biased,
and attempting interventions is rarely feasible.
As a result, learners may encounter fragments of knowledge,
but struggle to experience and connect the entire flow of thinking.
Keywords
Data Science Literacy
Human-Centered Data Science
Experiential Analytics
Observation
Simulation-Based Learning
Virtual World Learning
But what if learners could observe, interpret, and test their ideas not in the real world,
but safely inside a world that has been deliberately constructed?
How much deeper might their understanding of data science become?
By using data generated within a virtual world, we are creating a new way of learning—
one that allows learners to move seamlessly from analysis to intervention to evaluation within a single, coherent environment.
Research Achievements
S. Takahashi and A. Yoshikawa, "Data Science in an Agent-Based Simulation World," 2023 IEEE International Conference on Teaching, Assessment and Learning for Engineering (TALE), Auckland, New Zealand, 2023, pp. 1-6, doi: 10.1109/TALE56641.2023.10398326.
KAKENHI (C), 25K06648, 2025–2028. Design of Virtual-World Teaching Materials for Data Science Using Manga Cases and Agent-Based Simulation. PI: Satoshi Takahashi. JSPS.
Lead Researcher
Satoshi Takahashi