Using Machine Learning Models to Predict the Length of Stay in a Hospital Setting
| Autor: | Abdelahad Chraibi, Sondes Chaabane, Rachda Naila Mekhaldi, Sylvain Piechowiak, Patrice Caulier |
|---|---|
| Přispěvatelé: | Laboratoire d'Automatique, de Mécanique et d'Informatique industrielles et Humaines - UMR 8201 (LAMIH), Centre National de la Recherche Scientifique (CNRS)-Université Polytechnique Hauts-de-France (UPHF)-INSA Institut National des Sciences Appliquées Hauts-de-France (INSA Hauts-De-France) |
| Jazyk: | angličtina |
| Rok vydání: | 2020 |
| Předmět: |
Mean squared error
Standardization business.industry Hospital setting Computer science Data transformation (statistics) 02 engineering and technology Machine learning computer.software_genre Random forest [SPI.AUTO]Engineering Sciences [physics]/Automatic 03 medical and health sciences 0302 clinical medicine Open source Hyper parameters 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing [INFO]Computer Science [cs] 030212 general & internal medicine Gradient boosting Artificial intelligence business computer |
| Zdroj: | 8th World Conference on Information Systems and Technologies (WorldCIST’2020) 8th World Conference on Information Systems and Technologies (WorldCIST’2020), Apr 2020, Budva, Montenegro. pp.202-211 Trends and Innovations in Information Systems and Technologies ISBN: 9783030456870 WorldCIST (1) |
| Popis: | International audience; Proper prediction of Length Of Stay (LOS) has become increasingly important these years. The LOS prediction provides better services, managing hospital resources and controls their costs. In this paper , we compared two Machine Learning (ML) methods on the Microsoft available dataset. This data are been firstly preprocessed by combining data transformation, data standardization and data codification. Then, the Random Forest (RF) and the Gradient Boosting model (GB) were carried out, with a phase of hyper parameters tuning until setting optimal coefficients. Finally, the Mean Square Error (MAE), R-squared (R 2) and the Adjusted R-squared (Adjusted R 2) metrics are selected to evaluate model with parameters. The best model is saved to be trained with real data later using transfer learning techniques. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|