This repository contains supplementary materials for the following conference paper:
Yuma Miyazaki, Valdemar Švábenský, Yuta Taniguchi, Fumiya Okubo, Tsubasa Minematsu, and Atsushi Shimada.
E2Vec: Feature Embedding with Temporal Information for Analyzing Student Actions in E-Book Systems.
In Proceedings of the 17th International Conference on Educational Data Mining (EDM 2024).
https://doi.org/10.5281/zenodo.12729854
Convert EventStream log into Text file.
See Section 3.2 in the paper.
Train fastText with preprocessed text.
See Section 3.3 in the paper.
Make CodeBook for Aggregation.
Perform k-means++ clustering for action vectors.
See Section 3.4.1 in the paper.
Generate students vector in one lecture course.
Predict students grade with student vectors.
See Section 7 in the paper.
If you use or build upon the materials, please use the BibTeX entry below to cite the original paper (not only this web link).
@inproceedings{Miyazaki2024e2vec,
author = {Miyazaki, Yuma and \v{S}v\'{a}bensk\'{y}, Valdemar and Taniguchi, Yuta and Okubo, Fumiya and Minematsu, Tsubasa and Shimada, Atsushi},
title = {{E2Vec: Feature Embedding with Temporal Information for Analyzing Student Actions in E-Book Systems}},
booktitle = {Proceedings of the 17th International Conference on Educational Data Mining},
series = {EDM '24},
editor = {Benjamin Paaßen and Carrie Demmans Epp},
location = {Atlanta, GA, USA},
publisher = {International Educational Data Mining Society},
month = {07},
year = {2024},
pages = {434--442},
numpages = {9},
url = {https://educationaldatamining.org/edm2024/proceedings/2024.EDM-short-papers.42/2024.EDM-short-papers.42.pdf},
doi = {10.5281/zenodo.12729854},
}