Perbandingan Random Forest dan XGBoost dalam Penentuan Jenis Perawatan Pasien Berdasarkan Hasil laboratorium Darah

Trhecia Priscila Momongan; Jek Siang Jong; Halim Budi Santoso

doi:10.30865/json.v7i3.9378

Authors

Cia Universitas Kristen Duta Wacana
Jek Siang Jong Universitas Kristen Duta Wacana
Halim Universitas Kristen Duta Wacana

DOI:

https://doi.org/10.30865/json.v7i3.9378

Keywords:

Random Forest, XGBoost, Machine Learning, Ensemble Learning, laboratorium darah

Abstract

Umumnya, keputusan rawat inap dan rawat jalan seorang pasien ditentukan secara subyektif oleh dokter menggunakan hasil laboratorium darah. Pada pasien yang masuk melalui Instalasi Gawat Darurat, keputusan tersebut dilakukan oleh dokter umum yang menangani banyak pasien sekaligus sehingga berpotensi salah keputusan. Machine Learning dapat dipakai untuk membantu dokter dalam pengambilan keputusan jenis perawatan. Penelitian ini membandingkan performa Random Forest dan XGBoost dalam memprediksi jenis perawatan pasien. Dataset yang digunakan berisi 4.412 data pasien dari beberapa rumah sakit swasta di Indonesia, dengan 11 atribut mencakup profil pasien serta hasil laboratorium seperti hematokrit, hemoglobin, eritrosit, leukosit, trombosit, MCH, MCHC, dan MCV. Dataset termasuk dalam kategori ketidakseimbangan ringan karena terdiri dari 59,6% pasien rawat jalan dan 40,4% rawat inap. Pengujian dilakukan menggunakan parameter default maupun dengan pengaturan fitur. Hasil pengujian menunjukkan bahwa kinerja Random Forest lebih baik dibanding XGBoost. Model Random Forest terbaik didapat dengan menggunakan parameter default dengan feature selection, mencapai akurasi 77%, presisi 77%, recall 77%, F1-Score 76% dan ROC-AUC 81,4%. XGBoost terbaik diperoleh dengan menggunakan default parameter seluruh variabel, dengan akurasi 76%, presisi 74%, recall 76%, F1-Score 75% dan ROC-AUC 80,2%. Random Forest menunjukkan keseimbangan prediksi antar kelas dan sensitivitas serta spesifisitas tinggi sehingga efektif dalam mendukung keputusan klinis. Penelitian selanjutnya dapat difokuskan pada metode feature selection dan optimasi parameter yang mempertimbangkan waktu dan beban komputasi.

References

D. Dady, K. Ayu Erika, and R. Rachmawaty, “Support system in the making of nursing clinical decisions in the hospital room: a literature review,” Jurnal Ilmu Kesehatan, vol. 8, no. 2, pp. 173–181, 2020, doi: 10.1136/bmjopen-2023-080208.

M. K. Glette, “General practitioners decision making in questions of hospital admissions — a review of the literature,” in Safety and Reliability – Theory and Applications, Cepin and Bris, Eds., London: Taylor & Francis Group, 2017, pp. 2467–2472. doi: 10.1201/9781315210469-311.

T. Trinh, A. Elfergani, and M. Bann, “Qualitative analysis of disposition decision making for patients referred for admission from the emergency department without definite medical acuity,” BMJ Open, vol. 11, pp. 1–9, 2021, doi: 10.1136/bmjopen-2020-046598.

J. Riyono, C. E. Pujiastuti, D. Prayitno, A. Latifa, and R. Putri, “Decision tree for determining hospital treatment for COVID-19 patients based on hematology parameters using the C5.0 algorithm,” JISA (Jurnal Informatika dan Sains), vol. 07, no. 02, pp. 125–132, 2024, doi: https://doi.org/10.31326/jisa.v7i2.2103.

K. Paser, A. Salsabila, C. A. Gunawan, and H. Irawiraman, “Profil hematologi pasien malaria rawat inap di RSUD Panglima Sebaya Kabupaten Paser periode Januari 2015–Maret 2018,” JSK (Jurnal Sains dan Kesehatan), vol. 3, no. 4, pp. 551–557, 2021, doi: https://doi.org/10.25026/jsk.v3i4.535.

N. Adinda, I. Jaenuddin, and S. Surdianah, “Pemeriksaan darah lengkap pada sampel pasien dengan metode automatik dan metode Westergren di Balai Besar Laboratorium Kesehatan Makassar (BBLK Makassar) di Indonesia,” Filogeni(Jurnal Mahasiswa Biologi), vol. 4, no. 3, pp. 205–214, 2024.

Y. J. Simatupang and Rr. T. S. Hariyati, “Dampak penggunaan electronic health record terhadap tingkat kepuasan pasien: sebuah tinjauan literatur,” Jurnal Persatuan Perawat Nasional Indonesia (JPPNI), vol. 9, no. 2, pp. 103–115, Dec. 2024, doi: 10.32419/jppni.v9i2.536.

H. Crc and H. Informatics, Statistics and Machine Learning Methods for EHR Data. CRC Press, 2021. doi: 10.1201/9781003030003.

A. F. Nasution and P. A. Juledi, “Penerapan algoritma random forest untuk klasifikasi tingkat keparahan penyakit pada data rekam medis,” Jurnal of Computer Science and Information Systems (JCoInS), vol. 43, no. 3, pp. 47–58, 2025, doi: 10.36987/jcoins.v6i3.7993.

M. I. Sodikin, “Penerapan dan manfaat machine learning di rumah sakit,” Multiverse: Open Multidisciplinary Journal, vol. 2, no. 2, pp. 262–265, Aug. 2023, doi: 10.57251/multiverse.v2i2.1207.

A. Sindar Sinaga, S. Ramen, S. Mulyani, K. Kunci, M. Bayi, and E. Metode, “Teknik ensemble dalam machine learning untuk menentukan tingkat akurasi perkembangan motorik bayi 0–12 bulan,” Jurnal Saintikom (Jurnal Sains Manajemen Informatika dan Komputer), vol. 23, no. 2, pp. 418–425, Aug. 2024, doi: 10.53513/jis.v23i2.10059.

A. S. More and D. P. Rana, “Review of Random Forest classification techniques to resolve data imbalance,” in ICISIM(International Conference on Intelligent Systems and Information Management), 2017, pp. 72–78. doi: 10.1109/ICISIM.2017.8122151.

M. R. Ansyari, M. I. Mazdadi, F. Indriani, D. Kartini, and T. H. Saragih, “mplementation of random forest and extreme gradient boosting in the classification of heart disease using particle swarm optimization feature selection,” Journal of Electronics, Electromedical Engineering, and Medical Informatics, vol. 5, no. 4, pp. 250–260, 2023, doi: 10.35882/jeeemi.v5i4.322.

L. R. Sitompul, A. A. Nababan, M. L. Manihuruk, W. A. Ponsen, and S. Supriyandi, “omparison of XGBoost, Random Forest and Logistic Regression algorithms in stroke disease classification,” Sinkron, vol. 9, no. 2, pp. 957–968, 2025, doi: 10.33395/sinkron.v9i2.14794.

J. P. Anggraini, Chaya Gladys Zhafirah A, and A. Desiani, “Perbandingan algoritma Random Forest dan Extreme Gradient Boosting (XGBoost) dalam klasifikasi penyakit gagal jantung,” Komputika : Jurnal Sistem Komputer, vol. 14, no. 2, pp. 149–157, 2025, doi: 10.34010/komputika.v14i2.16618.

R. Harahap, M. Irpan, M. A. Dinata, L. Efrizoni, and Rahmaddeni, “Perbandingan algoritma Random Forest dan XGBoost klasifikasi penyakit jantung berdasarkan data medis,” Betrik, vol. 15, no. 02, pp. 130–141, 2024, doi: 10.37859/jf.v15i2.9927.

M. Linda, T. Alfianti, and R. Supriyanto, “Perbandingan kinerja algoritma Random Forest, AdaBoost, dan XGBoost dalam memprediksi risiko penyakit osteoporosis,” Ilmu Komputer Agri-Informatika, vol. 11, no. 2, pp. 173–183, 2024, doi: 10.29244/jika.11.2.172-184.

B. Septian Cahya Putra, I. Tahyudin, B. Adhi Kusuma, and K. Nur Isnaini, “Efektivitas algoritma Random Forest, XGBoost, dan Logistic Regression dalam prediksi penyakit paru-paru,” Jurnal Techno.COM, vol. 23, no. 4, pp. 909–922, Nov. 2024, doi: 10.62411/tc.v23i4.11705.

F. Aziz, A. Wahab, and P. Muna, “Penerapan metode machine learning untuk mengklasifikasi penanganan perawatan pasien,” JOPACS(Journal Pharmacy and Applicationof Computer Sciences), vol. 1, no. 2, pp. 57–62, 2023, doi: https://doi.org/10.59823/jopacs.v1i2.37.

D. B. Saputra, V. Atina, F. E. Nastiti, and F. I. Komputer, “Penerapan model CRISP-DM pada prediksi nasabah kredit menggunakan algoritma Random Forest,” Idealis: Indonesia Journal Information System, vol. 7, no. 2, pp. 240–247, Jul. 2024, doi: 10.36080/idealis.v7i2.3244.

A. N. Chary, A. Bhananker, B. Torres, V. S. Martell, M. N. Shah, and A. D. Naik, “Older adults’ experiences of emergency department admission decisions: a qualitative study in a public safety net hospital,” J. Gen. Intern. Med., vol. 40, no. 7, pp. 1567–1575, 2025, doi: 10.1007/s11606-024-09250-3.

W. E. Herter et al., “Impact of a machine learning-based decision support system for urinary tract infections: prospective observational study in 36 primary care practices,” JMIR Med. Inform., vol. 10, no. 5, pp. 1–15, 2022, doi: 10.2196/27795.

F. Orhan and M. N. Kurutkan, “redicting total healthcare demand using machine learning: separate and combined analysis of predisposing, enabling, and need factors,” BMC Health Serv. Res., vol. 25, no. 1, pp. 1–27, 2025, doi: 10.1186/s12913-025-12502-5.

W. Hong et al., “A comparison of XGBoost, Random Forest, and nomograph for the prediction of disease severity in patients with COVID-19 pneumonia: implications of cytokine and immune cell profile,” Front. Cell. Infect. Microbiol., vol. 12, no. April, pp. 1–13, 2022, doi: 10.3389/fcimb.2022.819267.

M. Sadikin, “EHR dataset for patient treatment classification,” Mendeley Data, 2020.

N. Maulidah, “Prediksi peningkatan jumlah nasabah deposito berjangka menggunakan algoritma KNN, Decision Tree, Random Forest dan XGBoost,” InComTech : Jurnal Telekomunikasi dan Komputer, vol. 13, no. 2, p. 90, 2023, doi: 10.22441/incomtech.v13i2.16921.

D. A. Murtiningsih, W. Sari, and I. N. Fajri, “Comparison of Light Gradient Boosting machine, Extreme Gradient Boosting, and CatBoost with balancing and hyperparameter tuning for hypertension risk prediction on clinical dataset,” JAIC(Journal of Applied Informatics and Computing), vol. 9, no. 5, p. 2753, 2025, doi: 10.30871/jaic.v9i5.10400.

E. Prasetyo, B. Prasetiyo, and P. Korespondensi, “Peningkatan akurasi klasifikasi algoritma C4.5 menggunakan teknik bagging pada diagnosis penyakit jantung,” JTIIK(Jurnal Teknologi Informasi dan Ilmu komputer), vol. 7, no. 5, pp. 1035–1040, Oct. 2020, doi: 10.25126/jtiik.202072379.

K. Aditya Ananta Wisnu Wardana and A. A. Mizwar Rahim, “Analisis perbandingan algoritma XGBoost dan algoritma Random Forest untuk klasifikasi data kesehatan mental,” LOGIC : Jurnal Ilmu Komputer dan Pendidikan, vol. 2, no. 5, pp. 808–818, Aug. 2024, doi: 10.36050/3v3xwn06.

D. T. Murdiansyah, “Prediksi stroke menggunakan Extreme Gradient Boosting,” JIKO (Jurnal Informatika dan Komputer), vol. 8, no. 2, p. 419, Sep. 2024, doi: 10.26798/jiko.v8i2.1295.

Y. Rombe, S. A. Thamrin, and A. Lawi, “Application of Adaptive Synthetic Nominal and Extreme Gradient Boosting Methods in Determining Factors Affecting Obesity: A Case Study of Indonesian Basic Health Research Survey 2013,” Indonesian Journal of Statistics and Its Applications, vol. 6, no. 2, pp. 309–317, 2022, doi: 10.29244/ijsa.v6i2p309-317.

A. Khairunnisa, “Perbandingan model Random Forest dan XGBoost untuk prediksi kejahatan kesusilaan di provinsi Jawa Barat,” JIKO (Jurnal Informatika dan Komputer), vol. 7, no. 2, p. 202, Sep. 2023, doi: 10.26798/jiko.v7i2.799.

R. D. Mendrofa, M. H. Siallagan, J. Amalia, and D. P. Pakpahan, “Credit risk analysis with Extreme Gradient Boosting and Adaptive Boosting algorithm,” Journal of Information System,Graphics, Hospitality and Technology, vol. 5, no. 1, pp. 1–7, Mar. 2023, doi: 10.37823/insight.v5i1.233.

Budi Sunarko, Agung Adi Firdaus, Yudha Andriano Rismawan, and Anan Nugroho, “Analysis of facial areas to identify CHD risks based on facial textures,” Jurnal Nasional Teknik Elektro dan Teknologi Informasi, vol. 14, no. 1, pp. 69–76, 2025, doi: 10.22146/jnteti.v14i1.13658.

M. H. Atan et al., “Automated detection of unattended children via deep learning,” Journal of Advanced Research in Applied Sciences and Engineering Technology, vol. 3, no. 3, pp. 11–20, 2026, doi: 10.37934/araset.62.3.1120.

J. Ben Temones, L. Domanais, J. C. De La Cruz, A. J. Timado, E. Codecio, and M. G. Llonado, “Machine learning-based teacher education student placement model via interest profile and diagnostic test,” Journal of Education Research, vol. 6, no. 1, pp. 223–232, 2025, doi: 10.37985/jer.v6i1.2331.

G. Sambasivam, G. Prabu kanna, M. S. Chauhan, P. Raja, and Y. Kumar, “A hybrid deep learning model approach for automated detection and classification of cassava leaf diseases,” Sci. Rep., vol. 15, no. 1, pp. 1–24, 2025, doi: 10.1038/s41598-025-90646-4.

E. Sastra Ompusunggu, A. Nainggolan, M. K. Sihombing, F. Sains, and D. Teknologi, “Penentuan kelayakan promosi pegawai menggunakan algoritma Random Forest classifier dan XGBoost classifier,” Jurnal TEKINKOM, vol. 6, no. 2, pp. 773–783, 2023, doi: 10.37600/tekinkom.v6i2.949.