Evaluation of student learning outcomes is a critical process in education that aims to measure the achievement of learning objectives. Through various methods such as tests, projects, and observations, teachers can assess students' understanding, skills, and progress in the subject matter. The purpose of applying data mining using the K-Means Clustering algorithm in evaluating student learning outcomes is to identify patterns that may be hidden in learning outcome data, divide students into groups based on their level of achievement or learning characteristics, and provide valuable insights to teachers and education stakeholders. The results of clustering student learning assessment data can uncover patterns that are beneficial to educators and school administrators. Analysis of these clusters can reveal information about achievement trends, trends in success or difficulty in specific subjects, as well as allow identification of students who need additional help. Grouping of cluster results based on student assessment data with k-means obtained 2 groups of students, namely Diligent students with group C0 and group students Very Diligent with group C1. The C0 group of Diligent students consists of 63 students and the C1 group consists of 91 Very Diligent students. The silhouette score test results for cluster 2 are as high as 0.9168 and show that grouping data into these groups is better, the use of silhouette score as an evaluation metric provides useful guidance in determining the optimal number of clusters in clustering analysis and data interpretation.
Copyrights © 2024