An Ensemble Model to Minimize Fluctuation Influences on Short-Term Medical Workload Prediction
Tasquia Mizan, Sharareh Taghipour · Scientia Iranica · 2022
Time series forecasting is an important field of machine learning since many real-world events are related to time. Real-time data are commonly prone to errors due to irregular fluctuations, seasonal biases, and missing values in the data. The erroneous data causes inaccurate forecasting which leads to business loss. Moreover, the concept drift problem is a known problem in time series forecasting that also results in poor forecasting accuracy. This work presents an Adaptive Batched-Ranked Ensemble (ABRE) model that reduces the effect of fluctuation using the time-variant windowing technique. A data aggregation technique is developed and integrated with the offline training phase of the proposed model to tackle the concept drift problem. A meta-model is developed from the offline phase. This meta-model is exposed in the online forecasting phase which ensures faster execution for incoming data. The model is implemented for the medical workload prediction after testing and comparing with a few other heterogeneous ensemble models. The comparison results show in terms of the root mean squared error, the proposed model performs at least 65.7% better than the heterogeneous stacked ensemble models applied to the experimental dataset. Moreover, the ABRE model reduces the prediction error by approximately 73.6%.