OPTIMASI SELEKSI FITUR PADA RANDOM FOREST UNTUK PREDIKSI RISIKO DIABETES TAHAP AWAL MENGGUNAKAN SFS, SBS, DAN PSO
DOI:
https://doi.org/10.32699/biner.v5i2.11298Keywords:
feature selection, Sequential Forward Selection, Sequential Backward Selection, Particle Swarm Optimization, Random Forest, diabetesAbstract
Preprocessing merupakan tahapan penting dalam machine learning karena kualitas data mempengaruhi performa model klasifikasi. Salah satu teknik preprocessing yang sering digunakan adalah feature selection untuk memilih atribut yang relevan dan mengurangi fitur yang kurang berpengaruh terhadap proses klasifikasi. Penelitian ini bertujuan melakukan optimasi seleksi fitur menggunakan metode Sequential Forward Selection (SFS), Sequential Backward Selection (SBS), dan Particle Swarm Optimization (PSO) pada algoritma Random Forest untuk prediksi risiko diabetes tahap awal. Dataset yang digunakan adalah Early Stage Diabetes Risk Prediction Dataset yang ditransformasikan ke bentuk numerik menggunakan Label Encoding. Hasil penelitian menunjukkan bahwa Random Forest tanpa seleksi fitur menghasilkan accuracy sebesar 98.27% dengan 16 fitur, metode SFS menghasilkan accuracy sebesar 98.65% dengan 11 fitur, sedangkan metode SBS memperoleh accuracy tertinggi sebesar 99.23% menggunakan 11 fitur dan PSO memperoleh accuracy sebesar 98.85% menggunakan 11 fitur. Selain meningkatkan accuracy, seleksi fitur juga mampu mengurangi jumlah atribut sehingga proses klasifikasi menjadi lebih efisien.
Downloads
References
[1] Mustofa, M. A. M., Sucipto, dan Nugroho, A. 2025. “Penerapan Random Forest untuk Deteksi Dini Penyakit Parkinson’s dengan Data Frekuensi Suara.” Jurnal Qua Teknika 15(2): 25–37.
[2] Sucipto, Prasetya, D. D., dan Widiyaningtyas, T. 2024. “Educational Data Mining: Multiple Choice Question Classification in Vocational School.” Matrik: Jurnal Manajemen, Teknik Informatika, dan Rekayasa Komputer 23(2): 379–388
[3] Triwardani, A. A., Daniati, E., dan Wardani, A. S. 2025. “Studi Komparatif Algoritma Machine Learning dalam Klasifikasi dan Prediksi Kanker Paru-Paru.” JMSI 6(2): 214–222.
[4] Guyon, I., dan A. Elisseeff. 2003. “An Introduction to Variable and Feature Selection.” Journal of Machine Learning Research 3: 1157–1182.
[5] Han, J., M. Kamber, dan J. Pei. 2012. Data Mining: Concepts and Techniques. 3rd ed. Waltham: Morgan Kaufmann.
[6] Islam, M. M. F., R. Ferdousi, S. Rahman, dan H. Y. Bushra. 2020. “Likelihood Prediction of Diabetes at Early Stage Using Data Mining Techniques.” Dalam Computer Vision and Machine Intelligence in Medical Image Analysis, 113–125. Singapore: Springer
[7] Kennedy, J., dan R. Eberhart. 1995. “Particle Swarm Optimization.” Dalam Proceedings of ICNN'95 - International Conference on Neural Networks, 1942–1948. IEEE.
[8] Nugroho, A., A. Z. Fanani, dan G. F. Shidik. 2021. “Wrapper Feature Selection for Random Forest Classifier Using Backward Elimination and Forward Selection for Diabetes Prediction.” Dalam 2021 International Seminar on Application for Technology of Information and Communication (iSemantic), 178–181. IEEE.
[9] Sokolova, M., dan G. Lapalme. 2009. “A Systematic Analysis of Performance Measures for Classification Tasks.” Information Processing and Management 45(4): 427–437.
[10] Breiman, L. 2001. “Random Forests.” Machine Learning 45(1): 5–32.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Biner : Jurnal Ilmiah Informatika dan Komputer
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
An author who publishes in this Journal agrees to the following terms:
- Author retains the copyright and grants the journal the right of first publication of the work simultaneously licensed under the Creative Commons Attribution-ShareAlike 4.0 License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal
- Author is able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book) with the acknowledgement of its initial publication in this journal.
- Author is permitted and encouraged to post his/her work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of the published work (See The Effect of Open Access).
