K-Means++ clustering of self-reported behavioral proxies yields six distinct, stable learner profiles (k = 6, Silhouette = 0.62)¶

Claim · All claims
Evidence · 1 study (2 entries) · associational r1–r2 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person · recheck by 2029: tool and model accuracy changes quickly

Subclaims¶

q2 i? Unsupervised clustering of 12 behavioral, psychological, and environmental features identified six learner profiles ranging from Highly Engaged Achievers to At-Risk Learners. → Najem 2026 q2 i? The clustering solution was stable across 10 independent runs (mean ARI = 0.94, SD = 0.03; centroid displacement < 0.02 normalized units). → Najem 2026 (2)

Evidence¶

Najem 2026¶

Najem, K., Zaoui Seghroucheni, Y., & Ziti, S. (2026). Behavioral clustering for adaptive learning object recommendation in higher education using learning analytics for personalized student success. Discover Education. https://doi.org/10.1007/s44217-026-02181-7

q2 · i? · associational · r2

Secondary analysis of a 14,003-record anonymized dataset using K-Means++ (k = 6) on 12 self-reported behavioral, psychological, and environmental inputs; the article reports "six distinct learner profiles" (C0–C5), with Silhouette = 0.62 described as acceptable for educational data mining.

"Unsupervised clustering techniques were applied to the preprocessed data set, which resulted in the identification of six distinct learner profiles, based on unique combinations of behavioral, psychological, and performance traits."

Najem 2026 (2)¶

Najem, K., Zaoui Seghroucheni, Y., & Ziti, S. (2026). Behavioral clustering for adaptive learning object recommendation in higher education using learning analytics for personalized student success. Discover Education. https://doi.org/10.1007/s44217-026-02181-7

q2 · i? · associational · r1

Stability analysis over 10 independent K-Means++ runs with different random seeds; the article reports "mean Adjusted Rand Index (ARI)" of 0.94 (SD = 0.03) and centroid displacement under 0.02 normalized units.

"Across runs, the mean Adjusted Rand Index (ARI) between pairwise cluster assignments was 0.94 (SD = 0.03), and the average centroid displacement was less than 0.02 normalized units, indicating high stability and reproducibility of the clustering solution"

Discussion¶

Related Claims¶

• Under the reported settings, K-Means++ achieved the best internal clustering indices, though the K-Means baseline used a single initialization — possibly the same claim (merge candidate)
• Cluster analysis identifies three stable learner profiles: Engaged and Literate, Uncritical Reliant, and Skeptical Minimal users — related
• K-means clustering of K–8 students' platform trace data yields seven distinct engagement profiles — related
• Exploratory K-means clustering identifies four heterogeneous student AI-user profiles differing in experience, competence, usefulness, trust, and control, but the solution's optimality and stability are not established — related
• Unsupervised hierarchical clustering of knowledge vectors recovers expert bundle structure without bundle labels — related
• Cluster analysis across both studies identifies three meta-attention knowledge profiles, including an overconfident group with a self-knowledge/strategy mismatch — related
• Affinity propagation clustering of gameplay attempts yielded an interpretable six-cluster solution (part.6)
• K-means clustering of standardized prior knowledge and final grades yields four performance profiles: Consistent Achievers, Growing Achievers, Declining Performers, and Struggling Performers — related
• Teachers' post-workshop platform engagement fell into three behavioral profiles, with most teachers showing minimal creation and browsing — related