Penerapan Algoritma K-Means Untuk Pengelompokan Siswa Sebagai Dasar Strategi Pembelajaran Pada Smk Negeri 4 Tanah Grogot

Fityan, Noor (2026) Penerapan Algoritma K-Means Untuk Pengelompokan Siswa Sebagai Dasar Strategi Pembelajaran Pada Smk Negeri 4 Tanah Grogot. S1 Teknik Informatika thesis, STMIK Widya Cipta Dharma.

	Text 2243032-S1-Jurnal.pdf Download (301kB)
	Text 2243032-S1-Teknik Informatika.pdf Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

Penelitian ini bertujuan untuk menerapkan algoritma K-Means dalam mengelompokkan siswa berdasarkan kemiripan nilai akademik sebagai dasar penyusunan strategi pembelajaran di SMK Negeri 4 Tanah Grogot. Data nilai akademik selama ini lebih banyak digunakan untuk pelaporan hasil belajar dan evaluasi sederhana sehingga belum dimanfaatkan secara optimal untuk mengetahui pola kemampuan akademik siswa. Penelitian menggunakan kerangka Cross Industry Standard Process for Data Mining (CRISP-DM) yang meliputi Business Understanding, Data Understanding, Data Preparation, Modelling, Evaluation, dan Deployment. Sampel penelitian terdiri atas 34 siswa kelas XI Program Keahlian Perhotelan Tahun Ajaran 2024/2025. Setelah proses pembersihan data, penggabungan mata pelajaran agama, transformasi data, dan seleksi fitur, diperoleh 10 fitur nilai mata pelajaran. Data dinormalisasi menggunakan Min-Max Normalization, kemudian dikelompokkan menggunakan algoritma K-Means. Jumlah cluster ditentukan secara otomatis melalui kombinasi Elbow Method dan Silhouette Score. Hasil penelitian memperoleh jumlah terbaik sebanyak enam cluster dengan nilai Silhouette Score sebesar 0,205. Distribusi anggota secara berurutan pada Cluster 1 sampai Cluster 6 adalah 7, 3, 7, 4, 8, dan 5 siswa. Nilai tersebut menunjukkan bahwa pemisahan antarkelompok belum kuat, tetapi hasil pengelompokan masih dapat memberikan gambaran pola kemiripan nilai akademik siswa. Hasil penelitian diimplementasikan dalam sistem berbasis website yang dapat mengelola data akademik, menjalankan proses clustering, serta menampilkan profil dan distribusi cluster. Keterbatasan utama penelitian ini adalah penggunaan nilai akademik sebagai satu-satunya dasar pengelompokan sehingga belum mencakup aspek keterampilan, sikap, kehadiran, keaktifan, dan kondisi siswa secara menyeluruh. ============================================================ This study aims to apply the K-Means algorithm to cluster students based on similarities in academic scores as a basis for developing learning strategies at SMK Negeri 4 Tanah Grogot. Academic score data have primarily been used for learning outcome reports and basic evaluations and therefore have not been optimally utilized to identify patterns in students’ academic abilities. The study employed the Cross Industry Standard Process for Data Mining (CRISP-DM), comprising Business Understanding, Data Understanding, Data Preparation, Modelling, Evaluation, and Deployment. The research sample consisted of 34 Grade XI students from the Hospitality Expertise Program in the 2024/2025 academic year. Following data cleaning, the merging of religious education subjects, data transformation, and feature selection, 10 academicscore features were obtained. The data were normalized using Min-Max Normalization and clustered using the K-Means algorithm. The number of clusters was determined automatically through a combination of the Elbow Method and Silhouette Score. The results identified six clusters as the optimal solution, with a Silhouette Score of 0.205. The respective membership distributions from Cluster 1 to Cluster 6 were 7, 3, 7, 4, 8, and 5 students. The score indicates that separation between clusters was not strong; nevertheless, the clustering results still described patterns of similarity in students’ academic scores. The findings were implemented in a website-based system that manages academic data, performs clustering, and displays cluster profiles and distributions. The main limitation of this study is its exclusive reliance on academic scores as the basis for clustering; therefore, it does not yet cover students’ skills, attitudes, attendance, participation, and overall circumstances.

Item Type:	Thesis (S1 Teknik Informatika)
Additional Information:	Pembimbing 1 : Wahyuni, S.Kom., M.Kom Pembimbing 2 : Rizki Galang Rahmadani, S.Kom., M.Kom
Uncontrolled Keywords:	K-Means, Clustering, CRISP-DM, Silhouette Score, Elbow Method.
Subjects:	Q Science > QA Mathematics > QA76 Computer software
Divisions:	Teknik Informatika
Depositing User:	Mr Fityan Noor
Date Deposited:	07 Aug 2026 05:47
Last Modified:	07 Aug 2026 05:47
URI:	http://repository.wicida.ac.id/id/eprint/6433

Actions (login required)

View Item