Clustering Of Student Learning Styles in the industri 4.0 Using KMeans Algorithm

Abstract

Clustering is a technique for grouping homogeneous data so that the points in each cluster are as similar as possible according to convenience measures such as Euclidean-based distance or correlation-based distance. In the industrial era 4.0, learning media, the environment, the way teachers teach will affect student learning styles. From research on learning styles, many researchers agree on the importance of identifying learning styles to accelerate their learning performance. The purpose of this study is to classify student learning styles in the industrial era 4.0 using the Kmeans algorithm and the elbow method. The research method used is a waterfall. The number of research subjects was 108 students. the results of the research on the number of clusters (K), namely 6, obtained cluster 1 as many as 27 students, cluster 2 as many as 24 students, cluster 3 as many as 21 students, cluster 4 as many as 17 students, cluster 5 as many as 11 students and cluster 6 as many as 8 students. The performance of the grouping results based on the silhouette coefficient is 0.302, which means the grouping structure is weak. In cluster 1, the highest number has auditory elements, followed by kinesthetic and visual elements. The development of ICT-based media is one of the factors of student learning styles in the industrial era 4.0