Unsupervised Clustering and Visual Learning Analytics for Intelligent Competency Assessment in Government Education Selection Systems

Artificial intelligence and learning analytics have become increasingly important in supporting data driven educational assessment and competency profiling. This study proposes an unsupervised machine learning approach for competency profiling of government education selection participants using clustering and learning analytics techniques based on Seleksi Kompetensi Dasar (SKD) assessment data. The dataset consisted of 2,150 participants and included four assessment variables, namely Tes Wawasan Kebangsaan (TWK), Tes Intelegensia Umum (TIU), Tes Karakteristik Pribadi (TKP), and total SKD score. Prior to clustering, all variables were standardized using the StandardScaler method. The K Means algorithm was employed to identify latent competency structures, while the optimal number of clusters was determined using the Elbow Method and Silhouette Score analysis. Clustering performance was evaluated using Silhouette Score, Davies Bouldin Index, and Calinski Harabasz Score. The experimental results demonstrated that the optimal clustering configuration was achieved at with a silhouette score of 0.585, indicating good cluster cohesion and separation. The findings revealed two distinct participant competency groups, where Cluster 0 represented high performing participants with consistently higher scores across all SKD dimensions, while Cluster 1 demonstrated lower competency readiness. Among all variables, TIU emerged as the most discriminative factor in competency differentiation. Furthermore, PCA and t SNE visualizations confirmed clear cluster separation with minimal overlap, indicating the presence of meaningful latent competency structures within the dataset. These findings demonstrate that unsupervised machine learning and learning analytics approaches can effectively support intelligent educational assessment, competency mapping, and data driven decision making in educational selection systems.

Authors

Institutions

Publication Details

Journal
Artificial Intelligence in Learning
Published
2026-11-17
DOI
https://doi.org/10.63913/ail.v2i4.75
Primary Topic
Online Learning and Analytics
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Unsupervised Clustering and Visual Learning Analytics for Intelligent Competency Assessment in Government Education Selection Systems

Chen Junru
Artificial Intelligence in Learning
Online Learning and Analytics
article

Unsupervised Clustering and Visual Learning Analytics for Intelligent Competency Assessment in Government Education Selection Systems

Chen Junru
article en

Abstract

Artificial intelligence and learning analytics have become increasingly important in supporting data driven educational assessment and competency profiling. This study proposes an unsupervised machine learning approach for competency profiling of government education selection participants using clustering and learning analytics techniques based on Seleksi Kompetensi Dasar (SKD) assessment data. The dataset consisted of 2,150 participants and included four assessment variables, namely Tes Wawasan Kebangsaan (TWK), Tes Intelegensia Umum (TIU), Tes Karakteristik Pribadi (TKP), and total SKD score. Prior to clustering, all variables were standardized using the StandardScaler method. The K Means algorithm was employed to identify latent competency structures, while the optimal number of clusters was determined using the Elbow Method and Silhouette Score analysis. Clustering performance was evaluated using Silhouette Score, Davies Bouldin Index, and Calinski Harabasz Score. The experimental results demonstrated that the optimal clustering configuration was achieved at with a silhouette score of 0.585, indicating good cluster cohesion and separation. The findings revealed two distinct participant competency groups, where Cluster 0 represented high performing participants with consistently higher scores across all SKD dimensions, while Cluster 1 demonstrated lower competency readiness. Among all variables, TIU emerged as the most discriminative factor in competency differentiation. Furthermore, PCA and t SNE visualizations confirmed clear cluster separation with minimal overlap, indicating the presence of meaningful latent competency structures within the dataset. These findings demonstrate that unsupervised machine learning and learning analytics approaches can effectively support intelligent educational assessment, competency mapping, and data driven decision making in educational selection systems.

Artificial Intelligence in LearningVol. 2(4)
University of Science and Technology Liaoning (CN)
Online Learning and Analytics
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon