Penerapan Algoritma K-Means untuk Identifikasi Jurusan Unggulan Berdasarkan Nilai Siswa (Studi Kasus: MTs Tarbiyatul Banin)

Riyan Hadi  Prabowo; Shasha Ramadhani Putri; Abdullah  Muttaqin; Atsna Rifqi Habibur  Rahman; Muhammad  Arifin

doi:10.37817/ikraith-informatika.v10i2.6861

Riyan Hadi Prabowo Universitas Muria Kudus
Shasha Ramadhani Putri Universitas Muria Kudus
Abdullah Muttaqin Universitas Muria Kudus
Atsna Rifqi Habibur Rahman Universitas Muria Kudus
Muhammad Arifin Universitas Muria Kudus

DOI: https://doi.org/10.37817/ikraith-informatika.v10i2.6861

Keywords: data mining, clustering, k-means, leading academic programs, subject scores, islamic junior high school

Abstract

This study was conducted to identify superior majors at MTs Tarbiyatul Banin Pekalongan, Pati, by applying the K-Means Clustering algorithm to grade data from 17 subjects across 226 students. The methodology consisted of data preprocessing including data cleaning and feature standardization, determination of the optimal number of clusters using the Elbow method which yielded three clusters, clustering using K-Means, and evaluation using the Silhouette Coefficient. The results revealed three performance clusters, namely the Superior category with the highest average grades, the Intermediate category with standard performance, and the Basic category with the lowest performance. The Silhouette Coefficient value obtained was 0.2208. Based on the proportion of students classified into the Superior category, the Sains Riset D Non Boarding major recorded the highest percentage at 53.12%, followed by Tahfidz Boarding A at 36.36%, and Sains Riset C Boarding at 31.25%. Therefore, Sains Riset D Non Boarding and Tahfidz Boarding A were objectively identified as superior majors due to their dominant concentration of high-achieving students.