Penerapan Data Mining Untuk Analisis Pola Kehadiran Serta Pengaruhnya Terhadap Prestasi Belajar Siswa Di SMA Negeri 7 Palembang

Abstract

Student absenteeism is not merely an administrative issue; it also reflects discipline and motivation to learn, which influence academic achievement. Unfortunately, in many schools, absenteeism data is merely recorded without further analysis. The researchers are interested in examining how data mining techniques can be used to uncover patterns of absenteeism and their relationship to academic achievement. This study was conducted at SMA Negeri 7 Palembang using data from 318 11th-grade students for the 2024/2025 academic year. The analysis involved data preprocessing, the creation of classification labels, and the application of the K-Nearest Neighbor (KNN) and Logistic Regression algorithms. The results showed that both algorithms were able to predict achievement categories with high accuracy, namely 95.83%. KNN with a value of K=3 was more balanced in identifying high-achieving students (F1-score 0.75), while Logistic Regression excelled in precision (1.00) but had lower recall (0.56). Regression analysis indicates that each additional absence reduces a student’s probability of achieving high academic performance by a factor of 0.74. These findings confirm that absences are not merely a number but a critical factor that can influence academic achievement. Therefore, schools are advised to utilize attendance and absence data not only for administrative purposes but also as an academic monitoring system so that learning strategies can be implemented more effectively.