Analisis Perbandingan Algoritma Linear Regression dan Polynomial Regression dalam Memprediksi Durasi Rawat Inap Pasien Rumah Sakit
Abstract
The length of stay (LoS) of hospital patients is an essential indicator for measuring the efficiency of healthcare services. Accurate LoS prediction helps hospitals optimize resource management, estimate costs, and improve service quality. This study compares the performance of Linear Regression and Polynomial Regression algorithms in predicting patient LoS. The dataset, obtained from Kaggle, consists of 835 patient records that underwent preprocessing and transformation. The independent variables include gender, age, disease type, and type of service, while LoS serves as the dependent variable. The research applies the Knowledge Discovery from Data (KDD) approach, which includes the stages of selection, cleaning, transformation, data mining, and evaluation. Experiments were conducted using three data-splitting ratios (70:30, 80:20, and 90:10) with evaluation metrics MAE, MSE, RMSE, and R². The results show that Linear Regression performed slightly better, with average R² values ranging between 0.18 and 0.20, indicating its potential to support hospital management efficiency.
Keywords: Length of Stay; Linear Regression; Polynomial Regression; Data Mining; Prediction.
Abstrak
Durasi rawat inap pasien (Length of Stay/LoS) merupakan indikator penting dalam mengukur efisiensi pelayanan rumah sakit. Prediksi LoS yang akurat membantu rumah sakit dalam pengelolaan sumber daya, estimasi biaya, dan peningkatan mutu layanan. Penelitian ini membahas perbandingan kinerja algoritma Linear Regression dan Polynomial Regression dalam memprediksi LoS pasien. Data penelitian diperoleh dari Kaggle dengan total 835 data pasien yang melalui proses preprocessing dan transformasi. Variabel independen meliputi gender, umur, jenis penyakit, dan jenis service, sedangkan LoS menjadi variabel dependen. Metode penelitian menggunakan pendekatan Knowledge Discovery from Data (KDD) yang mencakup tahapan selection, cleaning, transformation, data mining, dan evaluation. Pengujian dilakukan pada tiga rasio pembagian data (70:30, 80:20, dan 90:10) menggunakan metrik MAE, MSE, RMSE, dan R². Hasil menunjukkan Linear Regression memiliki performa sedikit lebih unggul dengan rata-rata R² sebesar 0,18–0,20, yang dapat mendukung efisiensi manajemen rumah sakit.
Keywords
References
A. Rezaianzadeh, M. Dastoorpoor, M. Sanaei, C. Salehnasab, M. Mohammadi, J. M, dan A. Mousavizadeh, “Predictors of length of stay in the coronary care unit in patient with acute coronary syndrome based on data mining methods”, Clinical Epidemiology and Global Health, vol. 8, no. 2, pp.383–388, Juni 2020.
M. Eskandari, A. H. Alizadeh Bahmani, H. A. Mardani-Fard, I. Karimzadeh, N. Omidifar, dan P. Peymani, “Evaluation of factors that influenced the length of hospital stay using data mining techniques”, BMC Med Inform Decis Mak, vol. 22, pp.1, Des 2022.
T. Kelley dkk., Interventions To Decrease Hospital Length of Stay Technical Brief Number 40 R, vol. 21-EHC015. Rockville: AHRQ, 2021. Diakses: 9 Oktober 2024. [Daring]. Tersedia pada: https://www.ncbi.nlm.nih.gov/books/NBK574435/pdf/ Bookshelf_NBK574435.pdf
M. Ba-Aoum, N. Hosseinichimeh, K. P. Triantis, K. Pasupathy, M. Sir, dan D. Nestler, “Statistical analysis of factors influencing patient length of stay in emergency departments,” International Journal of Industrial Engineering and Operations Management, vol. 5, no. 3, pp.220–239, Agu 2023, doi: 10.1108/ijieom-10-2022-0056.
S. Lam dkk., “Prediction of in-hospital mortality and length of stay using an early warning scoring system: clinical audit,” 2006. Diakses: 9 Oktober 2024. [Daring]. Tersedia pada: https://www.sciencedirect.com/science/article/pii/S1470211824006717
W. Chango, J.A. Lara, R. Cerezo, C. Romero, “A review on data fusion in multimodal learning analyticsand educational data mining”, WIREs Data Mining and Knowledge Discovery, vol. 12, no. 4, pp.1-19, April 2022
B. Wibisana, Y. Hendra Pratama, dan H. Purnomo, “Forecasting of Rice Production Using a Linear Regression and Polynomial Regression”, Journal of Informatics and Communications Technology, vol. 6, no. 1, pp.33–41, Juni 2024, doi: 10.52661.
K. I. Stone, R. I. Zwiggelaar, P. Jones, dan N. Mac Parthalá in, “A systematic review of the prediction of hospital length of stay: Towards a unified framework”, PLOS Digital Health, vol. 1, no. 4, pp.1-38, Apr 2022, doi: 10.1371/JOURNAL.PDIG.0000017.
M. Shahid Ansari dkk., “Identification of predictors and model for predicting prolonged length of stay in dengue patients”, Health Care Management Science, vol. 24, no. 4, pp.786–798, Des 2021, doi: 10.1007/s10729-021-09571-3.
K. Alghatani, N. Ammar, A. Rezgui, A. S. Nejad, “Predicting Intensive Care Unit Length of Stay and Mortality Using Patient Vital Signs: Machine Learning Model Development and Validation”, JMIR Publications Advancing Digital Health & Open Science, vol. 9, no. 5, pp.1–23, May 2023, doi: 10.2196/21347
A. Mehran, A. Eltahawi, M.H.A. Elaziz, M.N.A. Elwhab, “Predicting length of stay in hospitals intensive care unit using general admission features”, Ain Shams Engineering Journal, vol. 12, no. 4, pp.3691-3702, 2021
G. Xia, Y. Bi, dan C. Wang, “Optimization design of passive residual heat removal system based on improved genetic algorithm,” Annals of Nuclear Energy, vol. 189, Sep 2023, doi: 10.1016/j.anucene.2023.109859.
S. Lai, X. Hu, H. Xu, Z. Ren, dan Z. Liu, “Multimodal sentiment analysis: A survey”, Displays, vol. 80, Des 2023, doi: 10.1016/j.displa.2023.102563.
D. K. Sharma, M. Chatterjee, G. Kaur, dan S. Vavilala, “Deep learning applications for disease diagnosis”, Deep Learning for Medical Applications with Unique Data, hlm. 31–51, Jan 2022, doi: 10.1016/B978-0-12-824145-5.00005-8.
G. Romeo, “Data analysis for business and economics”, Elements of Numerical Mathematical Economics with Excel, pp. 695–761, 2020, doi: 10.1016/B978-0-12-817648-1.00013-X.
How To Cite This :
Refbacks
- There are currently no refbacks.
