Sep 2026· Modelling and Data Analysis· Vol 16, pp. 7-29· 0 citations· 12 references
TL;DR
MAE with a contrastive loss extracts latent student characteristics, not reducible to aggregated indicators, from big educational data from state examination data for categorizing students and supporting decision-making.
Abstract
Context and relevance. The digitalization of education generates large volumes of learning-activity data containing patterns inaccessible to traditional statistical methods limited to aggregated performance indicators. Deep learning can extract latent patterns from data structure. Objective. To develop and compare student’s vector representation approaches from examination results for cluster analysis and latent-pattern discovery. Hypothesis. Big data on examination results contain patterns not reducible to aggregated performance indicators, and neural-network methods can capture them as vector representations suitable for clustering. Methods and materials. Data from the GIA-9 and GIA-11 examinations for 2023–2025 were used (11 subjects, approximately 5,000 students and 500,000 interactions). Three approaches were compared: a baseline on aggregated scores (Score stats), a variational autoencoder (VAE), and a masked autoencoder (MAE), including a version trained with a contrastive loss. Clustering was performed with HDBSCAN after dimensionality reduction (PCA); quality was assessed with internal metrics (silhouette coefficient, DBCV) and t-SNE visualization. Results. Score stats yields a stable partition reflecting examination type, subject set, and performance. VAE achieves high internal-metric values under aggressive dimensionality reduction, but its clusters are determined by the interaction-sequence length. MAE without a contrastive loss does not produce meaningful clusters; with it, the metrics reach the Score stats level, and the clusters group students by characteristic combinations of jointly chosen examinations. Conclusions. MAE with a contrastive loss extracts latent student characteristics, not reducible to aggregated indicators, from big educational data. The approaches are applicable to state examination data for categorizing students and supporting decision-making.
The growing volume of educational data generated by Virtual Learning Environments (VLEs) presents unprecedented opportunities for data-driven student support. This paper proposes HELM a Hybrid Ensemble Learning framework with Multi-modal behavioral analytics designed to predict student academic performance and provide...
Lu-Yuan Li· Journal of Computing & B...· 0 citations
This study targets insufficient data resources and backward teaching modes in higher vocational Chinese education. It builds a three-dimensional data collection matrix covering classrooms, virtual communities and enterprise practice, gathering 12 data types from 2,000 students of five majors. OpenPose captures students...
Qing-Xiao Guan· International Conference on...· 0 citations
This review explains how explainability transforms predictions into interventions that faculty and students can trust and focuses on higher-level metrics beyond simply describing accuracy, such as area under the curve (AUC), F1 score, fairness metrics, and long-term learning outcomes, which helps ensure that both the t...
L. Ali· Dijlah Journal of Engineerin...· 0 citations
A temporal analysis model based on the fusion of Long Short-Term Memory networks and attention mechanisms to achieve personalized teaching path planning can effectively characterize the dynamic evolution of the learning process and provide accurate decision support for intelligent education systems.
Sheng-Jun Chen, Bingnv Luo, Yu-Ting Chen et al.· International Workshop on Ar...· 0 citations