Perbandingan Metode K-Means dan Metode DBSCAN pada Pengelompokan Saham Indeks LQ45

Article Sidebar

PDF
Published: Jul 14, 2026
DOI: https://doi.org/10.34128/jsi.v12i1.1319
Keywords:
Company Financial Performance, Clustering, DBSCAN, K-Means

Main Article Content

Iqbal Ali Mansyah Wardana
Institut Teknologi Sepuluh Nopember
Finda Nur Lailiyah
Institut Teknologi Sepuluh Nopember
Noviyanti Santoso
Institut Teknologi Sepuluh Nopember

Abstract

The increase in public interest in investment, especially among young people, has caused a surge in the number of investors in Indonesia since 2019. This condition is accompanied by the increasing number of companies listed on the Indonesia Stock Exchange (BEI), especially on the LQ45 stock index which is one of the main references for investors. . However, stock market dynamics often make it difficult for investors to choose shares with optimal financial performance. Financial ratio analysis of financial reports is an important tool for assessing company performance. However, financial ratio data is numerical and has not been clearly grouped, so a grouping method is needed to provide a better understanding of company performance patterns. This research compares two clustering methods, namely K-Means and DBSCAN, in grouping shares in the LQ45 index based on financial ratios such as Return on Equity (ROE), Net Profit Margin (NPM), Debt to Equity Ratio (DER), Debt to Asset Ratio (DAR), Company Size (UP), and Stock Trading Volume from the period February to July 2023. K-Means is known to be effective for clearly distributed data, while DBSCAN excels in handling data with complex distribution forms and the existence of outliers. The results of this research indicate that the DBSCAN method is better in grouping companies because it has higher Silhouette Score and Dunn Index values ​​compared to the K-Means method.

Article Details

How to Cite
[1]
I. A. M. Wardana, F. N. Lailiyah, and N. Santoso, “Perbandingan Metode K-Means dan Metode DBSCAN pada Pengelompokan Saham Indeks LQ45”, JSI, vol. 12, no. 1, pp. 80–89, Jul. 2026.
Issue
Vol. 12 No. 1 (2026): Jurnal Sains dan Informatika
Section
Articles
Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

Author Biographies

Iqbal Ali Mansyah Wardana, Institut Teknologi Sepuluh Nopember

Iqbal Ali Mansyah Wardana, lahir di Surabaya pada 18 Desember 2002. Saat ini menjadi mahasiswa aktif D4 Statistika Bisnis di Institut Teknologi Sepuluh Nopember Surabaya.  Bidang minat machine learning dan finance. Selain itu penulis juga menekuni di bidang optimasi portofolio.

Finda Nur Lailiyah, Institut Teknologi Sepuluh Nopember

Finda Nur Lailiyah, lahir di Surabaya pada tanggal 17 Februari 2003. Penulis merupakan mahasiswa aktif D4 Statistika Bisnis di Institut Teknologi Sepuluh Nopember Surabaya dan sedang menggeluti di bidang keuangan dan bidang optimasi portofolio.

Noviyanti Santoso, Institut Teknologi Sepuluh Nopember

Noviyanti Santoso, penulis mempunyai latar pendidikan bidang Statistika dan menekuni bidang Komputasi. Saat ini penulis menjadi staf pengajar di salah satu perguruan tinggi negeri di Surabaya. Selain itu penulis juga aktif dalam berbagai penelitian meliputi topik machine learning, text mining, dan big data analysis.

References

[1] A. Hidayah, “Investor Pasar Modal Tembus 10 Juta, Jebolan SMU Mendominasi,” CNBC Indonesia. Accessed: Jul. 23, 2026. [Online]. Available: https://www.cnbcindonesia.com/market/20221121103752-17-389764/investor-pasar-modal-tembus-10-juta-jebolan-smu-mendominasi

[2] S. H. Winarno, “Penilaian Kinerja Keuangan Perusahaan Melalui Analisis Profitabilitas,” Jurnal Moneter, pp. 106–112, 2017.

[3] C. Seniati and H. Firmansyah, “Indonesian Stock Movement Using K-Means Clustering Method,” Jurnal Sistem Informasi dan Teknologi Digital (SITEDI), vol. 2, no. 4, pp. 132–135, Dec. 2025, doi: 10.70888/sitedi.v2i4.70.

[4] R. A. Hasibuan, S. W. Rizki, and H. Perdana, “Perbandingan Metode Fuzzy Clustering Means dan Single Linkage pada Pengelompokan Saham LQ45,” Buletin Ilmiah Math. Stat. dan Terapannya (BIMASTER), vol. 10, no. 3, pp. 361–368, 2021.

[5] M. Shodikin and N. Taniasari, “Analisis Komparasi Algoritma DBSCAN dan K-Means dalam Pemetaan Segmentasi Pasien Rawat Inap Menggunakan Model RFMT,” MALCOM: Indonesian Journal of Machine Learning and Computer Science, vol. 5, no. 4, pp. 1231–1241, Oct. 2025, doi: 10.57152/malcom.v5i4.2218.

[6] S. Sulistina, P. E. P. Utomo, and B. F. Hutabarat, “Klasterisasi Wilayah Rawan Kriminalitas di Kota Jambi (2022-2024) Menggunakan Algoritma K-Means,” Jurnal INSTEK (Informatika Sains dan Teknologi), vol. 10, no. 2, pp. 546–560, Oct. 2025, doi: 10.24252/instek.v10i2.62051.

[7] N. P. E. Merliana, Ernawati, and Alb. J. Santoso, “Analisa Penentuan Jumlah Cluster Terbaik pada Metode K-Means Clustering,” in Proceeding SENDI_U, 2015, pp. 978–990.

[8] T. M. Kodinariya and P. R. Makwana, “Review on Determining Number of Cluster in K-Means Clustering,” International Journal of Advance Research in Computer Science and Management Studies, vol. 1, no. 6, pp. 90–95, 2013.

[9] R. Virgatama, A. Suprayogi, and H. S. Firdaus, “Identifikasi Pengaruh Sistem Keamanan Lingkungan Terhadap Tingkat Kejahatan Pencurian di Kota Surakarta dengan Metode Sistem Informasi Geografis,” Jurnal Geodesi Undip, vol. 8, no. 1, pp. 398–407, 2019.

[10] F. Setiawan, F. Azmi, and C. Setianingsih, “Clustering pada Data Sentimen Transportasi Online Menggunakan Algoritma DBSCAN,” e-Proceeding of Engineering, vol. 8, no. 5, pp. 6089–6096, 2021.

[11] U. Y. Purwanto, B. Barus, and H. A. Adrianto, Spatial Hotspots Clustering of Forest and Land Fires Using DBSCAN and ST-DBSCAN. Bogor: Repository IPB, 2012.

[12] I. Djajadi, IDX Stock Index Handbook v1.2. Jakarta: Indonesia Stock Exchange, 2021.