Application Of The K-Means Clustering Algorithm For Data Collection And Grouping Of Reading Monitoring In The Literacy Program In One Of The Public Elementary Schools In Bandung

Abstract

This study employs the K-Means Clustering algorithm within a data-gathering and monitoring framework for student literacy initiatives at a public primary school in Bandung. The research used a descriptive quantitative methodology, employing data from student reading activities, including the quantity of books read and reading comprehension scores. The K-Means algorithm analyzes this data to categorize children into three reading levels: high (Grade A), moderate (Grade B), and low (Grade C). The computation technique employs two variables ($x$ and $y$) denoting the number of books read and the corresponding comprehension scores. The study determined the final centroids for each cluster as follows: C1 (12.5, 88.75) for high ability, C2 (7.33, 75.0) for moderate ability, and C3 (3.0, 55.0) for poor ability. After two cycles, the clustering outcomes stabilized, indicating no further member reassignments among groups. Of the 10 students assessed, the algorithm categorized four into the high cluster, three into the intermediate cluster, and three into the low cluster. The findings indicate that the K-Means algorithm effectively classifies student literacy data in an objective, quantifiable manner. The execution of this algorithm helps educators track literacy progress, categorize abilities, and develop more targeted instructional strategies.