Abstract:We propose a novel data-driven machine learning method using long short-term memory (LSTM)-based multi-stage forecasting for influenza forecasting. The novel aspects of the method include the following: 1) the introduction of LSTM method to capture the temporal dynamics of seasonal flu and 2) a technique to capture the influence of external variables that includes the geographical proximity and climatic variables such as humidity, temperature, precipitation, and sun exposure. The proposed model is compared aga… Show more
“…The deep network architecture of the LSTM cells can provide a powerful model in temporal data processing. Recently, LSTM have attracted much interest in temporal data prediction of infection disease prediction such as in [22] where authors proposed a LSTM method to capture the temporal dynamics of seasonal u and for real-time in uenza forecasting.…”
Section: Long Short Term Memory Recurrent Neural Network (Lstm)mentioning
Due to the continuous spread of the novel coronavirus (COVID-19) worldwide, it is urgent to develop accurate decision-aided methods to support healthcare policymakers to control and early detect COVID-19 outbreak especially in the data science era. In this context, our main goal is to build a generic and accurate method that can predict daily conrmed cases which helps stake-holders to make and review their epidemic response plans. This method takes advantage of the complementarity of DNN (Deep Neuronal Networks), LSTM (Long Short Term Memory) and CNN (Convolutional Neuronal Networks) where their forecasted values represent the inputs of stacked ensemble meta-learners that will generate the nal outbreak predictions. To the best of our knowledge, this is the rst time that deep ensemble learning is used to deal with this issue. The proposed method is validated on three experimental scenarios, Tunisia case study, China case study and the third one is based on China data and models to predict Tunisia COVID-19 outbreak. Experiment results indicate that, compared with individual learners, the stacked-DNN meta-learner, whose input are forecasted values of DNN, LSTM and CNN, achieved the best accurate results in terms of accuracy as well as RMSE for the three scenarios. In conclusion, our ndings demonstrate that i) deep ensemble learning may be used as an accurate decision support tool for improving COVID-19 outbreak forecasting, ii) it is possible to reuse China learners and meat-learners to make prediction of the epidemic trend for other countries when preventive and control measures are comparable.
“…The deep network architecture of the LSTM cells can provide a powerful model in temporal data processing. Recently, LSTM have attracted much interest in temporal data prediction of infection disease prediction such as in [22] where authors proposed a LSTM method to capture the temporal dynamics of seasonal u and for real-time in uenza forecasting.…”
Section: Long Short Term Memory Recurrent Neural Network (Lstm)mentioning
Due to the continuous spread of the novel coronavirus (COVID-19) worldwide, it is urgent to develop accurate decision-aided methods to support healthcare policymakers to control and early detect COVID-19 outbreak especially in the data science era. In this context, our main goal is to build a generic and accurate method that can predict daily conrmed cases which helps stake-holders to make and review their epidemic response plans. This method takes advantage of the complementarity of DNN (Deep Neuronal Networks), LSTM (Long Short Term Memory) and CNN (Convolutional Neuronal Networks) where their forecasted values represent the inputs of stacked ensemble meta-learners that will generate the nal outbreak predictions. To the best of our knowledge, this is the rst time that deep ensemble learning is used to deal with this issue. The proposed method is validated on three experimental scenarios, Tunisia case study, China case study and the third one is based on China data and models to predict Tunisia COVID-19 outbreak. Experiment results indicate that, compared with individual learners, the stacked-DNN meta-learner, whose input are forecasted values of DNN, LSTM and CNN, achieved the best accurate results in terms of accuracy as well as RMSE for the three scenarios. In conclusion, our ndings demonstrate that i) deep ensemble learning may be used as an accurate decision support tool for improving COVID-19 outbreak forecasting, ii) it is possible to reuse China learners and meat-learners to make prediction of the epidemic trend for other countries when preventive and control measures are comparable.
“…Nasserie T., et al [10] used disease models such as the IDEA model to project influenza peaks and epidemic final sizes. Some researches [11][12][13] used long short-term memory (LSTM), a neural network model, to verify its effectiveness in flu prediction. Yang Wan.…”
Influenza causes numerous deaths worldwide every year. Predicting the number of influenza patients is an important task for medical institutions. Two types of data regarding influenza-like illnesses (ILIs) are often used for flu prediction: (1) historical data and (2) user generated content (UGC) data on the web such as search queries and tweets. Historical data have an advantage against the normal state but show disadvantages against irregular phenomena. In contrast, UGC data are advantageous for irregular phenomena. So far, no effective model providing the benefits of both types of data has been devised. This study proposes a novel model, designated the two-stage model, which combines both historical and UGC data. The basic idea is, first, basic regular trends are estimated using the historical data-based model, and then, irregular trends are predicted by the UGC data-based model. Our approach is practically useful because we can train models separately. Thus, if a UGC provider changes the service, our model could produce better performance because the first part of the model is still stable. Experiments on the US and Japan datasets demonstrated the basic feasibility of the proposed approach. In the dropout (pseudo-noise) test that assumes a UGC service would change, the proposed method also showed robustness against outliers. The proposed model is suitable for prediction of seasonal flu.
“…Some researchers used the number of patients in the past as features [9][10][11][12][13], while others integrated other data sources to predict the number of patients in the future. Examples of these sources are climatological data [14,15], search engine queries [16][17][18][19], public comments on social media like Twitter [20,21], online informationseeking behavior on websites like Wikipedia [22,23] and a combination of multiple data streams [15,[24][25][26]. Different methods on these features were applied.…”
Section: Introductionmentioning
confidence: 99%
“…Some researches treated the problem as an instance of more general time series forecasting using time series methods (ARIMA, ARIMA-STL, GARMA) [9,10,17,27], while others used ML methods including Stacked linear regression [24,26], AdaBoost regression with decision trees [26], GB [12], SVR [26,28], elastic net [28] RF [11,12,28], Artificial Neural Network (ANN) [12,20]. Recently, a DL method Called LSTM has attracted much interest in ILI prediction and gave excellent results, which are more accurate than those of other methods [12,13,15,29]. In addition to investigating the performance of the three different feature spaces with multiple time-series, ML and DL based methods to predict the weekly ILI rate in Syria; we proposed novel future spaces n − years − before_m − weeks − around that integrate into state-of-the-art ML and DL methods.…”
Objective: An accurate forecasting of outbreaks of influenza-like illness (ILI) could support public health officials to suggest public health actions earlier. We investigated the performance of three different feature spaces in different models to forecast the weekly ILI rate in Syria using EWARS data from World Health Organization (WHO). Time series feature space was first used and we applied the seven models which are Naïve, Average, Seasonal naïve, drift, dynamic harmonic regression (Dhr), seasonal and trend decomposition using loess (STL) and TBATS. The Second feature space is like some state-of-the-art, which we named 53 − weeks − before_52 − first − order − difference feature space. The third one, we proposed and named n − years − before_m − weeks − around (YnWm) feature space. Machine learning (ML) and deep learning (DL) model were applied to the second and third feature spaces (generalized linear model (GLM), support vector regression (SVR), gradient boosting (GB), random forest (RF) and long short term memory (LSTM)).
Results:It was indicated that the LSTM model of four layers with 1 − year − before_4 − weeks − around feature space gave more accurate results than other models and reached the lowest MAPE of 3.52% and the lowest RMSE of 0.01662. I hope that this modelling methodology can be applied in other countries and therefore help prevent and control influenza worldwide.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.