585-command expert-annotated Linux/bash dataset with balanced four-level representation¶
Product or Programme · All products and programmes
Evidence · 3 claims (3 for) · 1 study (1 design),q2· 0 of 1 report an effect size · 3 claims rest on one study
Description¶
A dataset created by Alonso-Carracedo et al. containing 585 Linux/bash commands, expert-annotated across four CogTax representation levels for evaluating automatic taxonomy-level classification.
Components¶
Claims¶
- An AST-only structural classifier substantially exceeds random baseline accuracy in predicting CogTax levels, with L1–L3 mutually confused and L4 perfectly precise [+W]
- A combined AST-plus-embedding classifier achieves 89% accuracy assigning CogTax levels to 585 expert-annotated Linux/bash commands, outperforming either representation alone [+W]
- Taxonomy level is defined as the maximum of cognitive complexity and operational impact, ensuring monotone coverage of both dimensions [+W]
Related Products and Programmes¶
-
Key Sources¶
- Alonso-Carracedo, M., Fernandez-Boullon, R., Celard, P., Rodríguez-Martínez, F. J., & Otero-Cerdeira, L. (2026). CogTax: A Four-Level Cognitive Taxonomy for Command-Line Computing Education. arXiv. https://arxiv.org/abs/2607.00140. doi:10.48550/arxiv.2607.00140
Page metadata
| Field | Value |
|---|---|
type |
product |
id |
585-command-expert-annotated-linuxbash-dataset-with-balanced-four-level-representation |
title |
585-command expert-annotated Linux/bash dataset with balanced four-level representation |
description |
A dataset created by Alonso-Carracedo et al. |
status |
review |
product_kind |
dataset |
generated |
by process:settle-candidates on 2026-10-10 |
trust |
unverified — no one has confirmed this page's content |
Frontmatter is the page's machine-readable record — see the Schema & Guide.