<![CDATA[Laboratory assistants play a crucial role in supporting computer science practicum learning; however, conventional evaluation methods based on agregate average scores are unable to identify hidden performance patterns and characteristics. This study aims to analyze the patterns and characteristics of laboratory assistant performance at STMIK Widya Cipta Dharma using an Educational Data Mining approach based on the K-Means clustering algorithm. This study employed a quantitative research method using questionnaire data collected from 178 student respo ndents with ten performance indicators measured using a five-point Likert scale. The analysis process consisted of several stages, including data collection, data preprocessing through data cleaning and attribute selection, instrument testing using Pearson Product Moment validity and Cronbach’s Alpha reliability analysis, Z-Score normalization to standardize indicator values, K-Means clustering using Euclidean distance calculations, evaluation of optimal cluster numbers using the Elbow Method, Silhouette Score, and Davies-Bouldin Index (DBI), and Principal Component Analysis (PCA) visualization for cluster interpretation. The results showed that the instrument was valid and reliable, and the K-Means algorithm produced three performance clusters consisting of Low Performance (20.8%), Medium Performance (55.1%), and High Performance (24.1%) groups.]]>
Copyrights © 2026
