A hybrid ARIMA-SVR approach for forecasting emergency patient flow
| العنوان: | A hybrid ARIMA-SVR approach for forecasting emergency patient flow |
|---|---|
| المؤلفون: | Li Luo, Dunhu Liu, Jianchao Yang, Ruixiao Kong, Yabing Feng, Yumeng Zhang |
| المصدر: | Journal of Ambient Intelligence and Humanized Computing. 10:3315-3323 |
| بيانات النشر: | Springer Science and Business Media LLC, 2018. |
| سنة النشر: | 2018 |
| مصطلحات موضوعية: | General Computer Science, Mean squared error, business.industry, Computer science, 020206 networking & telecommunications, Computational intelligence, 02 engineering and technology, Machine learning, computer.software_genre, Patient flow, Scheduling (computing), Support vector machine, Nonlinear system, 0202 electrical engineering, electronic engineering, information engineering, 020201 artificial intelligence & image processing, Autoregressive integrated moving average, Artificial intelligence, business, computer |
| الوصف: | The goal of this study is to explore and evaluate the use of a hybrid ARIMA-SVR approach to forecast daily radiology emergency patient flow. Owing to the fact that emergency patient flow is highly uncertain and dynamic, the forecasting problem is regarded as a complicated task. As the emergency patient flow may have both linear and nonlinear patterns, this paper presents a hybrid ARIMA-SVR approach, which hybridizes autoregressive integrated moving average (ARIMA) model and support vector regression (SVR) model to predict emergency patient arrivals. The proposed model is applied to 4 years of daily emergency visits data in the radiology department of a large hospital to justify the performance of the hybrid model against single models. The MAPE, RMSE and MAE of the hybrid model are 7.02%, 19.20 and 14.97, respectively. Furthermore, the hybrid model achieves better prediction performance than its competitors because it can capture the linear and nonlinear patterns simultaneously. Experimental results indicate that the proposed hybrid ARIMA-SVR approach is a promising alternative for forecasting emergency patient flow. These findings are beneficial for efficient patient flow management and scheduling decisions optimization. |
| تدمد: | 1868-5145 1868-5137 |
| URL الوصول: | https://explore.openaire.eu/search/publication?articleId=doi_________::4240207f068bcefd5d6ed3467139dac7 https://doi.org/10.1007/s12652-018-1059-x |
| حقوق: | CLOSED |
| رقم الأكسشن: | edsair.doi...........4240207f068bcefd5d6ed3467139dac7 |
| قاعدة البيانات: | OpenAIRE |
Find this article in full text from Springer Verlag. | تدمد: | 18685145 18685137 |
|---|