educational-data-mining
Filtering by topic educational-data-mining(3)Clear all filters
- PaperarXiv — Language & NLP (cs.CL)13 Jul 2026
A Durability and Cross-Language Transfer Benchmark for a Validated Teaching-Feedback Classification Protocol
Esteban U. Vega Barajas
The study tested a previously validated teaching-feedback classification protocol on Spanish and English datasets using three generations of language representations. The protocol showed durability, with a 2026 frontier model achieving the highest thematic F1 on Spanish but no sentiment advantage over simpler models. Model choice was found to be a deployment decision rather than a property of the method.
Original abstract
Institutions collect far more open-ended teaching-evaluation feedback than they read. A prior study introduced a validated protocol for classifying such comments by thematic category and sentiment, built from a documented annotation guide, an intra-annotator reliability measurement, stratified cross-validation, and a held-out evaluation on a Spanish institutional corpus with a frozen-encoder design. Two questions limit its reuse: whether a protocol fixed to 2019-era frozen embeddings stays competitive as representation methods advance, and whether it transfers to a second language. We re-run it on the original Spanish data across three representation generations, sparse lexical features, frozen transformer embeddings, and prompted large language models, and transfer its sentiment task to English with a balanced 45,000-comment corpus checked against an aspect-labeled education dataset. Treating paired comparisons as descriptive, we find the protocol durable: a 2026 frontier model posts the highest thematic F1 on the hardest Spanish task, yet shows no sentiment advantage over a cheap model and no descriptive separation from it on English, so model choice is a deployment decision, not a property of the method.
- PaperComputers and Education: Artificial Intelligence19 Jun 2026
Federated and explainable learning analytics for privacy-preserving academic risk modeling across heterogeneous educational institutions
William Villegas-Ch, Alexandra Maldonado Navarro, Jaime Govea, Joselin García-Ortiz et al.
This study proposes a federated, explainable learning analytics framework for modeling academic risk across institutions without sharing raw data. Experiments show that federated models maintain strong discriminative performance under heterogeneous conditions, while feature importance remains stable across institutions. The findings highlight the need for multidimensional evaluation in federated learning systems for education.
Original abstract
The increasing digitization of higher education has enabled the development of learning analytics models to identify students at risk of academic failure or dropout; however, most existing approaches rely on centralized training and assume homogeneous data distributions, limiting their applicability across institutions with heterogeneous student populations and interaction patterns, while privacy constraints restrict data sharing and hinder collaborative model development. To address these challenges, this study proposes a federated, explainable learning analytics framework for modeling academic risk trajectories under controlled institutional heterogeneity. The proposed architecture integrates temporal behavioral representations with socio-academic features within a multitask learning scheme, evaluated under both centralized and federated regimes, while modeling institutional heterogeneity through parameterized non-IID partitions that introduce controlled class imbalance, temporal drift, and feature-level variability. Experimental results show that federated models preserve strong discriminative performance and stable convergence as heterogeneity increases. At the same time, calibration metrics exhibit greater sensitivity to distributional shifts, revealing a decoupling between ranking performance and probabilistic reliability. In parallel, explainability analysis shows that the relative importance of features remains structurally stable across institutions, despite variations in contribution magnitudes. Cross-platform evaluation further shows that models retain discriminative capacity when transferred across educational environments, while exhibiting changes in calibration and explanatory intensity. These findings highlight the importance of multidimensional evaluation in federated learning systems, jointly considering performance, calibration, and interpretability, and provide a methodological framework for deploying robust and privacy-preserving learning analytics models in heterogeneous educational settings.
- PaperComputers and Education: Artificial Intelligence6 Jun 2026
Personalized neural cognitive architecture search: AutoML-driven diagnostic model generation for heterogeneous learner profiles
Liyuan Jia, Ke Dong
Proposed a personalized neural cognitive architecture search framework using automated machine learning to generate diagnostic models from multi-modal educational data. The framework integrates heterogeneous information networks for dynamic learner profiling and combines differentiable neural architecture search with Bayesian performance prediction. Experiments on 28.45 million records achieved 89.1% AUC, outperforming traditional psychometric and deep learning methods.
Original abstract
With the rapid advancement of digital transformation in education, personalized learning has become a central objective for improving educational quality. However, traditional educational models struggle to effectively process multi-source heterogeneous learning behavior data, resulting in limited learner profiling and inefficient diagnostic model development. In addition, existing teaching quality evaluation systems mainly rely on static examination outcomes and simple statistical indicators, lacking the capability to dynamically analyze learning processes. To address these challenges, this study proposes a personalized neural cognitive architecture search framework driven by automated machine learning. The framework integrates multi-modal educational data using heterogeneous information network technology to construct dynamic learner profiles incorporating cognitive states, behavioral sequences, and contextual factors. A hierarchical neural architecture search space is then designed, combining differentiable search strategies with a Bayesian performance predictor to automatically identify optimal diagnostic model structures. Experimental results based on 28.45 million educational records show that the proposed method achieves an AUC of 89.1%, accuracy of 85.3%, and F1-score of 84.7%. The model significantly outperforms traditional psychometric approaches and surpasses mainstream deep learning and AutoML baselines. Ablation experiments further confirm the importance of multi-source data fusion and automated architecture search. Despite strong performance, challenges remain regarding computational efficiency and fairness validation across diverse learner populations.