| Autor: |
Hitoshi Inoue, Koichi Yasutake, Osamu Yamakawa, Takahiro Tagawa, Takahiro Sumiya, Kiyomi Okamoto |
| Předmět: |
|
| Zdroj: |
International Conference on e-Learning; 2024, p321-323, 3p |
| Abstrakt: |
This study aims to clarify the multidimensional data structure in learning processes using a topological data analysis method, Mapper. Unlike conventional dimensionality reduction techniques such as Principal Component Analysis, Mapper visualizes data while preserving its inherent topological characteristics. The outcomes from Mapper significantly depend on selecting filter functions, covering parameter settings, and clustering algorithms. We conducted a series of experiments to evaluate these factors systematically using a comprehensive educational dataset. We present case studies demonstrating how different parameter configurations impact the analysis results, highlighting Mapper's potential in uncovering valuable insights from educational data. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
