Analysis of Students' Reading Interests Using the K-Means Clustering Algorithm

Abstract

This research aims to analyze students' reading interest using the K-Means Clustering algorithm. The purpose is to identify patterns in reading behavior and group students based on similarity in reading frequency, duration, and motivation. The dataset was collected from Kaggle and preprocessed through normalization and encoding. Using the K-Means algorithm with k = 3, the clustering process resulted in three distinct groups: low, medium, and high reading interest. The evaluation showed that most students fall into the moderate reading interest category. This study highlights the potential of clustering techniques in identifying behavioral patterns and can be used to improve campus literacy programs.