EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

A multi-dimensional hybrid stochastic model for early and interpretable blockage detection in programming education

Grota Abdelkader, Erritali Mohammed, Etcheverry Patrick, Nodenot Thierry

Discover Informatics · 21.08.2026 · 1 (1)

DOI: 10.1007/s44564-026-00007-0 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Discover Informatics, online erschienen am 21.08.2026. Verfasst von Grota Abdelkader und 3 weiteren Autor:innen. Die Arbeit nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Detecting student difficulties in programming education is a critical challenge for instructors, a task further complicated by the rise of AI coding assistants that can obscure the learning process. While fine-grained interaction logs from development environments offer rich behavioral data, existing analytical models often focus on a single dimension, limiting their predictive power and interpretability. This paper proposes a multi-dimensional hybrid stochastic model that integrates three complementary analytical layers to provide a more holistic understanding of student behavior. The model combines: (1) a Markov Chain to capture the behavioral dimension of workflow transitions; (2) a Hidden Markov Model (HMM) to infer the latent cognitive dimension, identifying states such as progression, hesitation, and blockage; and (3) a Recurrent Neural Network (RNN) with an attention mechanism to model the sequential dimension by capturing long-range temporal dependencies. We validated our approach on a dataset of interaction traces collected from 70 first-year computer science students. Empirical evaluation on 70 first-year students (student-level 5-fold cross-validation) reveals that all models, from a simple Random Forest to the full hybrid, achieve equivalent Macro-F1 of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\approx$$\end{document}90.7% on the 4-class classification task (Friedman \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\chi ^2(7) = 3.53$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$p = 0.83$$\end{document}, all pairwise Nemenyi comparisons non-significant). This equivalence is itself a contribution: it demonstrates that the behavioral taxonomy is sufficiently well-designed that any model can learn it. The hybrid model’s unique value lies in three properties that classification metrics do not capture: (1) multi-layer interpretability through HMM cognitive state labels, attention temporal emphasis, and Markov transition scores; (2) early detection with a mean lead time of 2.8 min before blockage becomes visible to the instructor; and (3) explicit uncertainty quantification through confidence-adaptive fusion weights. A pilot deployment with 12 instructors and 180 students confirmed that these properties translate into measurable instructional benefits: +18% exercises completed and +6.3 percentage points on final exams.

Zitieren

APA

Abdelkader, G., Mohammed, E., Patrick, E., & Thierry, N. (2026). A multi-dimensional hybrid stochastic model for early and interpretable blockage detection in programming education. Discover Informatics, 1(1). https://doi.org/10.1007/s44564-026-00007-0

Vancouver

Abdelkader G, Mohammed E, Patrick E, Thierry N. A multi-dimensional hybrid stochastic model for early and interpretable blockage detection in programming education. Discover Informatics. 2026;1(1). doi:10.1007/s44564-026-00007-0

BibTeX
@article{Abdelkader2026,
  author = {Abdelkader, Grota and Mohammed, Erritali and Patrick, Etcheverry and Thierry, Nodenot},
  title = {A multi-dimensional hybrid stochastic model for early and interpretable blockage detection in programming education},
  journal = {Discover Informatics},
  year = {2026},
  volume = {1},
  number = {1},
  doi = {10.1007/s44564-026-00007-0}
}
Kostenlos starten →