Data-driven Modeling as a Tool for Prediction of Future Outbreaks of Ebola Virus Disease in West Africa

Data-driven Modeling as a Tool for Prediction of Future Outbreaks of Ebola Virus Disease in West Africa | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Data-driven Modeling as a Tool for Prediction of Future Outbreaks of Ebola Virus Disease in West Africa Thomas James Wanyama This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6969004/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Ebola is among the deadliest human diseases and some measures to combat it have been developed, but they are still not able to fully minimize its far-reaching effects. However, the integration of data-driven modeling into existing Ebola prevention and control measures offers hope toward combating this viral disease. Decision trees (DT), gradient boosting (GB), deep neural networks (DNN), and a hybrid model were used to predict Ebola cases and deaths in Guinea, Liberia, and Sierra Leone from 2017 to 2026. These models were evaluated using mean absolute scaled error (MASE), root mean squared error (RMSE), and coefficient of determination (R²). The uncertainties in the prediction of the models were also quantified using the bootstrapping method. Of all the models, the DT model had higher feature importance scores for cases in two countries. The DT model also had a higher accuracy and better performance by RMSE and R 2 in predicting Ebola cases and deaths than other models. Predictions of Ebola cases and deaths by the DNN model increased from 2017 to 2026 with higher cases and deaths in 2026 while predictions of Ebola deaths by the hybrid model increased from 2017 to 2026 with higher deaths in 2026. The projected Ebola cases and deaths were higher in Sierra Leone than in other countries. These findings portray the likely number of cases and deaths in case an Ebola outbreak in the three mentioned countries. Furthermore, they show their significance in predicting Ebola virus disease and also have the possibility to help decision-makers in designing effective decisions for the early detection of Ebola incidents. The results of this study show that the DT and DNN models perform better than the other models on the collected Ebola virus disease dataset in the three countries. Therefore, the integration of data-driven infectious disease modeling approaches such as DT and DNN with intervention scenarios such as vaccination can altogether help to reduce the predicted number of Ebola cases and deaths in Guinea, Sierra Leone, and Liberia in the face of an outbreak. Decision Tree Deep Neural Network Ebola virus Disease Gradient Boosting Data-driven Modeling Uncertainty quantification Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Figure 14 Figure 15 Figure 16 Figure 17 Figure 18 1. Background Ebola virus disease (EVD) is a complex zoonosis that has been reported as being highly virulent in humans [ 1 ]. It is among the deadliest viral diseases in human history due to its high death rates [ 2 ]. EVD is initiated by the Ebola virus (EBOV) [ 3 ]. Some of the symptoms of EVD are: initial fever, fatigue before descending into headaches, vomiting, violent diarrhea, multiple organ failure, and massive internal bleeding [ 4 ]. Globally, one of the regions that is affected by EVD is Sub-Saharan Africa, since it has continually battled this fatal disease with frequent outbreaks [ 5 ]. For example, the 2014 Ebola outbreak is reported to have claimed 11306 deaths out of 28200 cases in 10 West African countries [ 6 ]. Some of these countries include: Guinea, Liberia, Mali, Nigeria, Senegal, and Sierra Leone. This 2014 Ebola outbreak held a greater magnitude than all the other previous outbreaks [ 7 ]. This West African outbreak had never been reported previously in the region despite Ebola being known to cause outbreaks in central and eastern Africa [ 8 ]. Some of the reasons behind the prevalent nature of the West African outbreak include the highly mobility of the West African communities and their gigantic population [ 8 ]. Other factors include climate change [ 9 ], urbanization, and the rising demand for food that is got from animals [ 10 ]. There have been efforts to handle any EVD outbreak that may occur up in the near future [ 5 ]. However, these efforts may not be fully effective as they may present some back and forth challenges. Therefore, relying on smart measures such as data-driven modeling is imperative. This data-driven modeling encompasses the use of forecasting models which utilize maximum data for reducing bias and permit automatic selection of predicting variables [ 11 ]. It can serve as an effective early warning tool in disease prediction. For example, data-driven modeling approaches such as machine learning have been used as effective tools to predict the mortality risk of COVID-19 patients [ 12 ]. Another study reported the use of tree-based machine learning (ML) algorithms and feed-forward neural networks as effective tools for early predictions of Kyasanur forest disease, an emerging zoonotic disease [ 13 ]. Statistical models such as logistic regression and ML models such as support vector machines, random forests, and decision trees have been used in the prediction of COVID-19 [ 14 ]. Time-series deep learning models such as the improved Long Short-Term Memory deep learning (DL) method have also been used in Iran, Russia, and Peru to predict the epidemic trends of COVID-19 [ 15 ]. Other than the improved Long Short-Term Memory DL method, other Time-series DL models such as Recurrent Neural Network (RNN) and Back Propagation Neural Network (BPNN) have been used in the prediction of Hepatitis B in China [ 16 ]. In the present study, regression-based models, DL models, and a hybrid model were utilized to predict EVD cases and deaths in Guinea, Liberia, and Sierra Leone for the next 10 years. To the best of the author’s knowledge, this is the first report that has used DT, GB, DNN, and a hybrid model (built from ANN, DT, and GB) to predict an Ebola outbreak in all the three mentioned countries. The three countries were selected because of the high Ebola cases and deaths recorded compared to other countries in West Africa (Nigeria, Senegal, and Mali), where Ebola cases and deaths were minimal. Despite immense studies on Ebola outbreak in the three countries [ 17 ], [ 18 ], [ 19 ], [ 20 ], [ 21 ], [ 22 ] and with other studies covering the disease’s forecasting, none of these studies have considered forecasting of both Ebola cases and deaths using multiple models involving tree-based ML models, DL models, and a hybrid model. Yet having a comprehensive study involving multiple forecasting models is key since different models capture patterns differently. Regression-based ML algorithms were chosen over time-series forecasting approaches due to their ability to process non-linear and complex data, which is a big limitation for time series models [ 23 ]. Regression algorithms are also comparatively insensitive to co-variation among the input predictors and are non-parametric thus offering them an advantage over time series models [ 13 ]. 2. Methods This section presents information on the dataset, and the steps done to prepare the data for forecasting. A summary of this section is shown in Fig. 1 . 2.1 Data collection The dataset was obtained from an online repository accessible at https://www.kaggle.com/datasets/imdevskp/ebola-outbreak-20142016-complete-dataset . This dataset contained data on the EVD outbreak in West Africa recorded from August 2014 to March 2016 (Table 1 ). Table 1 Dataset and its features S/n Dataset Name Features in the dataset 1 Ebola | 2014–2016 | Western Africa Ebola Outbreak Country 2 Year of reporting 3 Cumulative number of Ebola cases (confirmed, probable and suspected cases) 4 Cumulative number of Ebola deaths 2.2 Data pre-processing This step involved filtering and modifying the data so that it is easy to explore and understand. For example, the dataset that was obtained from the online repository contained all the data for the three countries. However, for easy exploration, data for each country were obtained from the overall dataset and then saved as an independent file. Additionally, the columns on cumulative number of confirmed, probable, and suspected cases and deaths were renamed as Ebola cases and deaths, respectively. Data normalization was also carried out. 2.3 Feature engineering A feature is a characteristic of any object. In machine learning, a sample is disintegrated into a set of features before training and testing for tasks, such as classification, prediction, or clustering [ 24 ]. In the current study, feature engineering was used to extract relevant features (data on year, Ebola cases, and deaths) from the dataset. 2.4 Learning algorithms The algorithms used were decision trees (DT), gradient boosting (GB), deep neural networks (DNN), and a hybrid model (built from decision trees, gradient boosting, and artificial neural networks). A brief description of these algorithms is given below. Gradient boosting ; this is a powerful and efficient learning method that is based on an implementation of gradient-boosted decision trees, and it is mostly deployed for supervised learning tasks, mainly regression and classification problems [ 25 ]. As an iterative approach, gradient boosting creates numerous trees in a sequential order of successive decision trees [ 14 ]. It has been complemented with a feature selection process that increases the overall performance by extracting the most relevant features from the input data [ 26 ]. Decision trees ; Decision trees have been utilized in many applications such as classification, regression, and feature selection [ 28 ]. The idea behind decision trees is that they recursively divide the data into subsets based on the values of different attributes until a stopping criterion is met [ 29 ]. The end-result is a tree-like structure, where each node represents a decision or a split based on a specific attribute [ 30 ]. This tree structure makes it possible for users to understand and interpret the decision-making process easily [ 29 ]. While using decision trees, a sequence of Boolean decisions are made with the aim to group data into categorical bins [ 31 ]. Decision trees also represent a mapping relationship between object attributes and object values, and their basic process follows the "divide and conquer" strategy [ 32 ]. Deep neural networks ; deep neural networks are part of the neural networks that are built to mimic the biological structures of the human brain [ 14 ]. They work best in the over-parameterized regime, with many more parameters than data points [ 34 ]. They are composed of units called neurons, which are connected to each other in a network [ 14 ]. Artificial neural networks ; these were developed to copy the features of the biological neurons in the human brain and the nervous system, and they are able to keep the biological concept of artificial neurons [ 36 ]. They consist of initial input data, activation function, and producing output with an output function, while the activation function can provide a smooth transition as input values modify [ 37 ]. The ANN is composed of connections, while each connection is indicated by a weight as its related importance, which can provide the output of one neuron as the input of another neuron [ 38 ]. In the Ebola forecasting modeling, the historical incidence is used as the input neurons, while the related predicting incidence is obtained from the output neurons after the ANN is well trained. 2.5 Computation of feature importance As a key step in data-driven modeling, feature importance (FI) helps to provide an understanding of how decisions are made and how this enhances the establishment of true causality between vital data attributes and outcomes in model inference [ 40 ]. In this study, FI scores were calculated to reflect the input features for the four models [ 24 ] since different models are known to produce different FI values due to variations in their learning algorithms [ 41 ]. Year and cases were taken as the input features while the target was death. 2.6 Model hyper-parameter tuning Prior to training the data, hyper-parameters were carefully configured so as to optimize the performance of the different algorithms. This process particularly involved adjusting the number of trees, learning cycles, learning rate, size of layers, and the number of training epochs. 2.7 Training of algorithms Training of the algorithms was done using Matlab R2024a software. It involved splitting the data into the training and testing sets were done. 80% of the data were incorporated into the training set, and 20% added to the testing set. 2.8 Data Visualization Graphical representations of the predicted data on EVD cases and deaths were performed on Matlab R2024a ( http://www.mathworks.com/ ), and this helped to identify trends and patterns in the EVD outbreak. 2.9 Evaluation Criteria Evaluation of the algorithms was done using three performance metrics. These were; mean absolute scaled error (MASE), root mean squared error (RMSE), and coefficient of determination (R²) (Fig. 7 ). These statistics helped to compare the target and output values, and also calculate a score as an index for the performance and accuracy of the forecasting algorithms [ 42 ]. 2.10 Uncertainty Quantification The processes of ML and DL have diverse sources of uncertainty such as vagueness due to extrapolation, variance in model parameters, inherent noise in data, and appropriateness of model selection [ 43 ]. Quantifying these underlying uncertainties is crucial as it helps to establish trust, determine risk in alternatives, or communicate the potential for error [ 43 ]. In the present study, uncertainty quantification (UQ) was done using the bootstrapping method to explore the confidence in the predictions by decision trees, gradient boosting, deep neural networks, and the hybrid model. 3. Results This section presents the historical data on Ebola cases and deaths from 2013 to 2016 in Guinea, Liberia, and Sierra Leone. It goes ahead to present the likely number of EVD cases and deaths in the next ten years from when the EVD outbreak was last recorded in the three West African countries. 3.1 Historical Ebola cases and deaths According to the year, the mean Ebola cases and deaths were extremely high in 2015 (i.e. 12,714 cases and 3,786 deaths), nearly twice the next highest mean cases in 2014 (Table 2 ). The mean EVD cases and deaths from 2013 to 2016 by country were high in Sierra Leone (i.e. 11,596 cases and 3,432 deaths) compared to other countries (Table 3 ). In Guinea, the historical Ebola cases and deaths were extremely high in 2014 and then underwent a constant increase from 2014 to 2015, followed by a sharp increase from 2015 to 2016. In Liberia, the historical Ebola cases and deaths were immensely high in 2014 and then underwent a constant increase from 2014 to 2015, followed by a sharp increase from 2015 to 2016. In Sierra Leone, the historical Ebola cases and deaths were tremendously high in 2014 and then underwent a constant increase from 2014 to 2015, followed by a sharp increase from 2015 to 2016. Table 2 The historical EVD cases and deaths by country and year, including the mean, minimum, and maximum of the cases and deaths Historical Ebola cases Historical Ebola deaths Country Year Mean Min-Max Mean Min-Max Guinea 2014 1741.842 648–2707 1064.789 430–1709 2015 3564.200 2730–3810 2363.959 1739–2536 2016 0 0 0 0 Liberia 2014 5659.946 1871–8018 2565.703 1089–3423 2015 6906.315 4-10666 3084.552 1-4806 2016 3560 5-10666 1604.333 3-4806 Sierra Leone 2014 5060.289 1026–9446 1366.974 422–2758 2015 12713.7 9633–14122 3786.064 2827–3955 2016 0 0 0 Table 3 Summary of the historical case and death counts for Ebola Viral Disease by country, including the mean, minimum, and maximum of the cases and deaths. Historical Ebola cases Historical Ebola deaths Country Mean Min-Max Mean Min-Max Guinea 3297.75 648–3810 2174.01 430–2536 Liberia 6752.04 4-10666 3019.61 1-4806 Sierra Leone 11596.25 1026–14122 3431.79 422–3955 3.2 Feature importance results Overall, cases as a feature input had higher FI scores than year for the four models, implying that variations in Ebola cases are strongly predictive of Ebola deaths in Guinea, Liberia, and Sierra Leone (Table 4 ). The higher FI scores for the year feature in Guinea using the gradient boosting model reveal strong time-based patterns, such as increasing or reducing Ebola deaths over time in Guinea than in other countries. Overall, the decision tree model had higher FI scores for cases in two countries compared to other models (Table 4 ). Table 4 FI scores for the different models Country Input features DT GB DNN Hybrid model Guinea Year 0 8.94934E-06 0.000473691 0.000534377 Cases 1 0.999991051 0.999526322 0.999465623 Liberia Year 0 3.38267E-05 4.89429E-06 1.9976e-05 Cases 1 0.999966173 0.999995112 0.999980024 Sierra Leone Year 0.001029601 0.001291345 0.000134385 0.000781894 Cases 0.998970399 0.998708655 0.999865592 0.999218106 3.3 Visualization of EVD case and death forecasts in the three countries Visualization of EVD case and death forecasts from decision trees Prediction using DT revealed a constant increase in Ebola cases and deaths in Guinea, Liberia, and Sierra Leone from 2017 to 2026 (Figs. 7 , 8 , 9 ). The predicted Ebola case counts were higher than the Ebola death counts in the three countries. Among the three countries, higher Ebola case and death counts (about 14000 cases and 4000 deaths) were registered in Sierra Leone. Guinea followed second with higher case and death counts (about 3800 cases and 2500 deaths) than Liberia. Visualization of EVD case and death forecasts from gradient boosting Forecasting using GB showed a constant increase in Ebola cases and deaths in all the countries from 2017 to 2026 (Figs. 10 , 11 , 12 ). Additionally, the predicted Ebola case counts were higher than the Ebola death counts in the three countries. Of all the three countries, higher Ebola case and death counts (about 14000 cases and 4000 deaths) were recorded in Sierra Leone. Liberia followed second with higher case counts (about 7000 cases) than Guinea. However, the death counts in Guinea and Sierra Leone were similar (about 3000 deaths). Visualization of EVD case and death forecasts from deep neural networks Prediction using DNN revealed a continual increase in Ebola cases and deaths in all the countries from 2017 to 2026 (Figs. 13 , 14 , 15 ). The projected Ebola case counts were higher than the Ebola death counts in the three countries. Extremely high Ebola case and death counts (about 14000 cases and 3900 deaths) were recorded in Sierra Leone. This was followed by Liberia (approximately 4000 cases and 3800 deaths), and then Guinea (approximately 3750 cases and 2500 deaths). Visualization of EVD case and death forecasts from the hybrid model The hybrid model revealed a sharp increase in the Ebola cases and deaths in Guinea and Sierra Leone from 2017 to 2026 (Figs. 16 and 18 ). However, in Liberia, Ebola cases were anticipated to gradually increase while the Ebola death counts were shown to gradually decrease from 2017 to 2026 (Fig. 17 ). The prediction also revealed higher Ebola case and death counts in Sierra Leone than in other countries. 3.4 Performance of the forecasting models based on the country Decision trees, gradient boosting, deep neural networks, and the hybrid model were used as the forecasting algorithms to predict the next Ebola cases and deaths in Guinea, Liberia, and Sierra Leone. Performance of the decision tree model Based on the country, the decision tree model performed well by R 2 and MASE in predicting Ebola cases and deaths in Liberia (Table 5 ). By RMSE, the model had a higher accuracy in predicting Ebola cases and deaths in Guinea than in other countries. Table 5 Performance metrics of the decision trees by country Cases Deaths Country MASE RMSE R 2 MASE RMSE R 2 Guinea 19.75 380.96 0.74 25.01 258.88 0.76 Liberia 0.89 4590.79 0.01 0.87 2050.45 0.01 Sierra Leone 22.11 1566.52 0.75 16.18 388.17 0.83 Performance of gradient boosting model This model had a good accuracy and good performance by RMSE and R 2 respectively in predicting Ebola cases and deaths in Guinea than in other countries (Table 6 ). By MASE, the model performed poorly in predicting Ebola cases and deaths than the naïve forecasts in all the three countries. Table 6 Performance metrics of gradient boosting by country Cases Deaths Country MASE RMSE R 2 MASE RMSE R 2 Guinea 4.57 377.66 0.61 4.83 260.92 0.83 Liberia 1.12 4715.79 0.01 1.11 2108.44 0.01 Sierra Leone 5.26 1873.38 0.53 4.86 505.03 0.61 Performance of deep neural networks The DNN model had the best accuracy by RMSE in predicting Ebola cases and deaths in Guinea than in other countries while a good performance in the prediction by R 2 was observed in Sierra Leone (Table 7 ). Table 7 Performance metrics of deep neural network by country Cases Deaths Country MASE RMSE R 2 MASE RMSE R 2 Guinea Inf 385.98 0.73 Inf 271.21 0.74 Liberia Inf 4610.21 0.00 Inf 2059.30 0.00 Sierra Leone Inf 1590.76 0.74 Inf 401.80 0.82 Performance of the hybrid model By R 2 and MASE, the hybrid model performed better in predicting Ebola cases and deaths in Liberia than in other countries (Table 8 ). A good accuracy by RMSE in predicting Ebola cases and deaths was also registered in Guinea than other countries. Table 8 Performance metrics of the hybrid model by country Cases Deaths Country MASE RMSE R 2 MASE RMSE R 2 Guinea 73.09 1143.32 -1.33 89.00 761.38 -1.08 Liberia 0.77 3826.11 0.31 0.76 1695.02 0.32 Sierra Leone 70.40 4085.15 -0.70 67.40 1208.70 -0.65 3.5 Overall model performance All models performed poorly in predicting Ebola cases and deaths compared to the naïve forecast. By RMSE and R 2 , the decision tree model had a higher accuracy and performance better in predicting Ebola cases and deaths than other models (Table 9 ). Table 9 Performance of the forecasting models MASE RMSE R 2 MASE RMSE R 2 Cases Deaths Decision trees 14.25 2179.42 0.50 14.02 899.17 0.53 Gradient boosting 3.65 2322.28 0.38 3.60 958.13 0.48 Deep neural networks Inf 2195.65 0.49 Inf 910.77 0.52 Hybrid model 48.09 3018.20 -0.57 52.39 1221.70 -0.47 3.6 Uncertainty quantification results by model Decision tree ; there was a greater uncertainty in the prediction of Ebola cases in 2017 and 2018 by decision trees. There was a high confidence in the prediction of Ebola cases from 2019 to 2026 by decision trees (Table 10 ). This model also predicted higher Ebola cases (3565) from 2019 to 2026, having a 95% confidence interval of 3530 to 3601. Gradient boosting ; there was a greater uncertainty in the prediction of Ebola cases in 2017 and 2018. A high confidence in the prediction of Ebola cases and deaths from 2019 to 2026 and 2017 to 2026 respectively was also registered (Table 10 ). The model predicted higher Ebola cases (3566) and deaths (2365) from 2019 to 2026 having a 95% confidence interval of 3530 to 3603 and 2340 to 2392 respectively. Deep neural network ; there was a greater uncertainty in the prediction of Ebola cases and deaths from 2017 to 2026. Predictions of Ebola cases and deaths also increased from 2017 to 2026 with higher cases (3556) and deaths (2406) in 2026 having a 95% confidence interval of 1013 to 6102 and 749 to 4117 respectively (Table 10 ). Hybrid model ; there was a high confidence in the prediction of Ebola deaths from 2022 to 2026. A greater uncertainty in the prediction of Ebola deaths from 2017 to 2021, 2024 to 2026 was also recorded. Predictions of Ebola deaths by this hybrid model increased from 2017 to 2026 with higher deaths (1590) in 2026 having a 95% confidence interval of 1034 to 2157 (Table 10 ). Table 10 Uncertainty Quantification using the bootstrapping method (CI: Confidence interval) Ebola cases Ebola deaths Decision trees Gradient boosting Deep Neural Network Hybrid Model Decision trees Gradient boosting Deep Neural Network Hybrid Model Year Mean 95% CI Mean 95% CI Mean 95% CI Mean 95% CI Mean 95% CI Mean 95% CI Mean 95% CI Mean 95% CI 2017 1737.76 1570.56, 1913.64 1737.76 1570.56, 1913.64 2415.08 -521.62, 5232.57 0.00 0, 0 0.00 0, 0 1062.24 959.97, 1173.23 1461.01 − 576.5, 3388.73 841.08 163.62, 1501.42 2018 1737.76 1570.56, 1913.64 1737.76 1570.56, 1913.6 2587.71 -417.57, 5566.83 0.00 0, 0 0.00 0, 0 1062.24 959.97, 1173.23 1617.34 -366.27, 3557.95 893.19 228.72, 1539.18 2019 3565.06 3529.7, 3601.37 3566.15 3530.41, 3602.59 2784.49 -456.02, 6080.08 0.00 0, 0 0.00 0, 0 2365.33 2339.81, 2391.62 1795.61 − 209, 3922.62 1386.98 713.92, 2098.54 2020 3565.06 3529.7, 3601.37 3566.15 3530.41, 3602.59 3071.06 163.1, 6110.50 0.00 0, 0 0.00 0, 0 2365.33 2339.81, 2391.62 1970.35 177.54, 3848.0 1445.23 848.37, 2072.16 2021 3565.06 3529.7, 3601.37 3566.15 3530.41, 3602.59 3298.85 714.12, 5778.31 0.00 0, 0 0.00 0, 0 2365.33 2339.81, 2391.62 2117.95 470.43, 3756.18 1494.43 945.84, 2042.02 2022 3565.06 3529.7, 3601.37 3566.15 3530.41, 3602.59 3494.60 2478.33, 4476.06 0.00 0, 0 0.00 0, 0 2365.33 2339.81, 2391.62 2295.80 1675.64, 2975.66 1553.71 1344.58, 1779.75 2023 3565.06 3529.7, 3601.37 3566.15 3530.41, 3602.59 3542.57 2690.79, 4185.72 0.00 0, 0 0.00 0, 0 2365.33 2339.81, 2391.62 2358.69 1915.2, 2763.27 1574.67 1425.88, 1705.65 2024 3565.06 3529.7, 3601.37 3566.15 3530.41, 3602.59 3528.57 1723.65, 5331.34 0.00 0, 0 0.00 0, 0 2365.33 2339.81, 2391.62 2384.24 1190.65, 3526.94 1583.19 1188.47, 1965.41 2025 3565.06 3529.7, 3601.37 3566.15 3530.41, 3602.59 3532.23 1347.01, 5708.40 0.00 0, 0 0.00 0, 0 2365.33 2339.81, 2391.62 2395.18 970.64, 3826.47 1586.83 1109.47, 2063.22 2026 3565.06 3529.7, 3601.37 3566.15 3530.41, 3602.59 3556.08 1013.24, 6102.24 0.00 0, 0 0.00 0, 0 2365.33 2339.81, 2391.62 2406.74 748.75, 4116.53 1590.69 1034.21, 2156.95 Confidence intervals are represented as (x, y) where x indicates the lower interval and y indicates the upper interval 4. Discussion 4.1 Ebola predictions in Guinea, Liberia, and Sierra Leone The present findings show that in 2014 and from 2015 to 2016, Ebola cases and deaths were exceptionally high in Guinea, Liberia, and Sierra Leone. This may be attributed to factors such as intense deforestation, which are known to increase the human-animal contact, thereby raising the risk of zoonotic disease transmission [ 44 ]. Furthermore, the immense population growth and urbanization, which have enhanced the encroachment on the habitats of wildlife, can be blamed [ 45 ]. With the aid of data-driven modeling, Ebola cases and deaths were forecasted to rise from 2016 to 2026 in case of an EVD outbreak in the three countries. However, extremely high Ebola cases and deaths were projected to occur in Sierra Leone than in other countries. The possible rise in Ebola cases and deaths from 2016 to 2026 may be attributed to enormous anthropogenic influences such as deforestation. This is possible as deforestation alters the movement patterns and populations of host species [ 46 ]. Due to deforestation which destroys wildlife habitats, animals such as bats are forced to move into human-inhabited regions formerly unfamiliar to wildlife [ 47 ], thus fuelling zoonotic outbreaks such as Ebola. This has also been reported in another study where increased spread of Ebola was linked to deforestation [ 48 ]. Other studies have also reported that pathogens such as viruses are easily transferred from animals to humans, especially if they are around deforested lands [ 49 ]. The urgent need to elevate industrial development in many countries has also driven the destruction of natural habitats for wildlife, thus increasing possible outbreaks of zoonotic diseases [ 50 ]. The high rate of cross-boundary movements between countries where Ebola has been previously reported may also lead to high Ebola cases and deaths in the future. For example, the recent Ebola outbreaks that have been recorded in the Democratic Republic of Congo, Uganda, and Guinea [ 51 ] prove it that cross boundary movements are significant contributors of Ebola outbreaks. The rise in Ebola cases and deaths in Guinea, Sierra Leone, and Liberia from 2017 to 2026 may also be linked to changes in global temperatures which are known to influence the emergence of zoonotic diseases such as Ebola [ 52 ]. 4.2 Characteristics of forecasting models Of all the models used, it was the decision trees and deep neural networks that performed better in predicting Ebola cases and deaths. This performance was based on model evaluation and UQ results. In another study, neural networks and decision trees outperformed other forecasting models in predicting COVID-19 cases [ 14 ]. Other similar studies reported a better performance of neural networks in predicting Parkinson’s disease [ 53 ] and Alzheimer’s disease [ 54 ] than other forecasting models. Another study reported neural network models as performing better than the other models in predicting Monkeypox outbreaks in the USA, Germany, the UK, France, and Canada [ 39 ]. The results of the present study where deep neural networks had a better performance after decision trees, are in agreement with a study where neural networks outperformed other models in forecasting [ 55 ]. 4.3 Limitations of the study, risks of misinterpretation, and integration of intervention scenarios Much as the predictions of Ebola cases and deaths from the four forecasting models are paramount, these predictions can only be relied on in case measures to combat EVD outbreaks are kept constant. This is because measures such as banning of travel to infected communities and early vaccination in infected or suspected communities can help to minimize Ebola cases and deaths. However, in some parts of the world and more so in low-and middle-income countries where resources needed to fully combat zoonotic outbreaks such as Ebola are extremely minimal, the predictions from the current study can be trusted. The findings in this study do not necessarily imply that an outbreak of Ebola should be expected in Guinea, Sierra Leone, and Liberia. However, the findings give a clue to the likely number of cases and deaths in case an Ebola outbreak is reported in the three countries. Integrating intervention scenarios such as vaccination campaigns with data-driven infectious disease modelling approaches can altogether help to reduce the predicted number of Ebola cases and deaths in Guinea, Sierra Leone, and Liberia in the face of an outbreak. 5. Conclusion and future direction Ebola is one of the deadliest diseases in human history, as it causes significant indirect and direct losses in the communities where it occurs. This underlines the pressing need for timely uncovering, quick reactions, and appropriate public health involvement to combat their spread and effects not only in Guinea, Sierra Leone, and Liberia but also other African countries where outbreaks are likely. Currently, no vaccine for Ebola is fully effective. It is also not easy to accurately predict how severe EVD could be. Due to this, the present research designed a hybrid model using an EVD dataset and compared it with the DT, GB, and DNN models. The time series dataset was obtained from the three mentioned countries. In predicting Ebola cases and deaths, the DT model gave relatively good RMSE (2179.42, 899.17) and R-squared values (0.50, 0.53), and this was followed by the DNN model with RMSE (2195.65, 910.77) and R-squared values (0.49, 0.52). Integrating exogenous covariates such as mobility, demographics, and climate will be pivotal in enhancing forecasting realism. This is because these exogenous covariates are associated with the transmission of zoonotic diseases such as Ebola. Despite the worth of the findings of the present study, future research should focus on the incorporation of data-driven modeling and SIR/SEIR models, as this can help to enhance the present standard epidemiology compartmental models in terms of accuracy. Abbreviations EVD Ebola virus disease ML Machine Learning DT Decision Tree GB Gradient Boosting DNN Deep Neural Network ANN Artificial Neural Network FI Feature importance CI Confidence Interval SIR Susceptible-Infectious-Recovered SEIR Susceptible-Exposed-Infectious-Recovered UQ Uncertainty Quantification Declarations Ethical approval Not applicable. Consent to participate Not applicable. Consent to publish This manuscript has the consent of the author. Funding statement This study did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Author Contribution The research concept and design, methodology, drafting of the manuscript, and critical revision of the manuscript for important intellectual content were all done by Thomas James Wanyama. Acknowledgement I would like to thank Fortunate Amanya for the support and courage provided during the course of this study. Data Availability The raw data that was used for predictive modelling is available at [https://www.kaggle.com/datasets/imdevskp/ebola-outbreak-20142016-complete-dataset](https:/www.kaggle.com/datasets/imdevskp/ebola-outbreak-20142016-complete-dataset) . References Pigott DM, et al. Mapping the zoonotic niche of Ebola virus disease in Africa. Elife. 2014;3:e04395. Ghareeb OA. Ebola-a fatal emerging zoonotic disease: a review. Ann Rom Soc Cell Biol. 2021;25(6):8748–54. Jacob ST, et al. Ebola virus disease. Nat Rev Dis Prim. 2020;6(1):13. Tambo E, Ugwu EC, Ngogang JY. Need of surveillance response systems to combat Ebola outbreaks and other emerging infectious diseases in African countries. Infect Dis poverty. 2014;3:1–8. Lokko CNM, Lokko FO. EPIDEMIOLOGY OF HUMAN EBOLA VIRUS DISEASE OUTBREAK IN SUB-SAHARAN AFRICA, 2025. Lokko CNAAM, Lokko FO, HUMAN EBOLA VIRUS DISEASE OUTBREAK IN SUB SAHARAN AFRICA. : IMPLICATION FOR BORDER TOWNS ACROSS THE GLOBE, 2016. Alexander KA, et al. What factors might have led to the emergence of Ebola in West Africa? PLoS Negl Trop Dis. 2015;9(6):e0003652. Coltart CEM, Lindsey B, Ghinai I, Johnson AM, Heymann DL. The Ebola outbreak, 2013–2016: old lessons for new epidemics. Philos Trans R Soc B Biol Sci. 2017;372(1721):20160297. White RJ, Razgour O. Emerging zoonotic diseases originating in mammals: a systematic review of effects of anthropogenic land-use change. Mamm Rev. 2020;50(4):336–52. WHO. Regional Office for Africa. In Africa, 63% jump in diseases spread from animals to people seen in last decade, 2024. Hashem S, et al. Machine learning prediction models for diagnosing hepatocellular carcinoma with HCV-related chronic liver disease. Comput Methods Programs Biomed. 2020;196:105551. Chowdhury MEH, et al. An early warning tool for predicting mortality risk of COVID-19 patients using machine learning. Cognit Comput. 2024;16(4):1778–93. Murthy RK. Early detection and prediction of zoonotic disease events using event-based surveillance and machine learning. Washington State University; 2023. Chen S, Ding Y. The feasibility of using machine learning to predict COVID-19 cases. Int J Med Inf. 2025;196:105786. https://doi.org/10.1016/j.ijmedinf.2025.105786 . Wang P, Zheng X, Ai G, Liu D, Zhu B. Time series prediction for the epidemic trends of COVID-19 using the improved LSTM deep learning method: Case studies in Russia, Peru and Iran. Chaos Solitons Fractals. 2020;140:110214. Xia Z, Qin L, Ning Z, Zhang X. Deep learning time series prediction models in surveillance data of hepatitis incidence in China. PLoS ONE. 2022;17(4):e0265660. McNamara LA. Ebola Surveillance—Guinea, Liberia, and Sierra Leone. MMWR Suppl, 65, 2016. Shaman J, Yang W, Kandula S. Inference and forecast of the current West African Ebola outbreak in Guinea, Sierra Leone and Liberia. PLoS Curr. 2014;6:ecurrents–outbreaks. Meltzer MI, et al. Estimating the future number of cases in the Ebola epidemic—Liberia and Sierra Leone, 2014–2015. MMWR Surveill Summ. 2014;63:1–14. Liu A. Diagnosis and Prognosis Prediction for Ebola Virus Disease Using Machine Learning Methods, 2021. Mannan A, Shoket N, Rahman JU, Uwitije R. Dynamic analysis of ebola virus disease with non-linear incidence rate using morlet wavelet neural networks and hybrid optimization techniques. Model Earth Syst Environ. 2025;11(2):79. Dou Y, Meng W. Predictive Analysis of The Number of Ebola Virus Infections Based on Grey Prediction Model and Neural Network, in 2022 IEEE 10th Joint International Information Technology and Artificial Intelligence Conference (ITAIC) , IEEE, 2022, pp. 190–195. Sreenivasan MA, Bhat HR, Rajagopalan PK. The epizootics of Kyasanur Forest disease in wild monkeys during 1964 to 1973. Trans R Soc Trop Med Hyg. 1986;80(5):810–4. Kumar A, Saxena N, Jung S, Choi BJ. Improving detection of false data injection attacks using machine learning with feature selection and oversampling. Energies. 2021;15(1):212. Wiens M, Verone-Boyle A, Henscheid N, Podichetty JT, Burton J. A tutorial and use case example of the eXtreme gradient boosting (XGBoost) artificial intelligence algorithm for drug development applications. Clin Transl Sci. 2025;18(3):e70172. Upadhyay D, Manero J, Zaman M, Sampalli S. Gradient boosting feature selection with machine learning classifiers for intrusion detection on power grids. IEEE Trans Netw Serv Manag. 2020;18(1):1104–16. Ye L, Jabbar S, Abdul Zahra M, Tan ML. Bayesian Regularized Neural Network Model Development for Predicting Daily Rainfall from Sea Level Pressure Data: Investigation on Solving Complex Hydrology Problem, Complexity , vol. 2021, pp. 1–14, Apr. 2021. 10.1155/2021/6631564 Wang C, Xu J, Tan S, Yin L. Secure decision tree classification with decentralized authorization and access control. Comput Stand Interfaces. 2024;89:103818. Mienye ID, Jere N. A survey of decision trees: Concepts, algorithms, and applications. IEEE access. 2024;12:86716–27. Rivera-Lopez R, Canul-Reich J, Mezura-Montes E, Cruz-Chávez MA. Induction of decision trees as classification models through metaheuristics. Swarm Evol Comput. 2022;69:101006. Itzkin M, Palmsten ML, Buckley ML, Birchler JJ, Torres-Garcia LM. Developing a decision tree model to forecast runup and assess uncertainty in empirical formulations. Coast Eng. 2025;195:104641. https://doi.org/10.1016/j.coastaleng.2024.104641 . Ji XC, et al. Recent advances in machine learning for defects detection and prediction in laser cladding process. Next Mater. 2025;7:100404. https://doi.org/10.1016/j.nxmate.2024.100404 . Atiku S, Obagbuwa I. Machine Learning Classification Techniques for Detecting the Impact of Human Resources Outcomes on Commercial Banks Performance, Appl. Comput. Intell. Soft Comput. , vol. 2021, pp. 1–16, Sep. 2021. 10.1155/2021/7747907 Mingard C, Rees H, Valle-Pérez G, Louis AA. Deep neural networks have an inbuilt Occam’s razor. Nat Commun. 2025;16(1):220. Rajesh G, Raajini XM, Dang H. Industry 4.0 interoperability, analytics, security, and case studies. CRC; 2021. Peterson DA, Sejnowski TJ. A dynamic circuit hypothesis for the pathogenesis of blepharospasm. Front Comput Neurosci. 2017;11:11. Mehlig B. Machine learning with neural networks: an introduction for scientists and engineers. Cambridge University Press; 2021. Cao W, Wang X, Ming Z, Gao J. A review on neural networks with random weights. Neurocomputing. 2018;275:278–87. Manohar B, Das R. Artificial neural networks for the prediction of monkeypox outbreak. Trop Med Infect Dis. 2022;7(12):424. Arrieta AB, et al. Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI. Inf fusion. 2020;58:82–115. Rengasamy D, et al. Feature importance in machine learning models: A fuzzy information fusion approach. Neurocomputing. 2022;511:163–74. Ardabili SF, Mahmoudi A, Gundoshmian TM. Modeling and simulation controlling system of HVAC using fuzzy and predictive (radial basis function, RBF) controllers. J Build Eng. 2016;6:301–8. Jalaian B, Lee M, Russell S. Uncertain context: Uncertainty quantification in machine learning. AI Mag. 2019;40(4):40–9. Figueroa D, Duprat X. Remedying anthropogenic zoonoses. Anim Sentience. 2020;5:29. Tian H et al. Urbanization prolongs hantavirus epidemics in cities, Proc. Natl. Acad. Sci. , vol. 115, no. 18, pp. 4707–4712, 2018. Rochlin I, Egizi A, Lindström A. The original scientific description of the lone star tick (Amblyomma americanum, Acari: Ixodidae) and implications for the species’ past and future geographic distributions. J Med Entomol. 2022;59(2):412–20. Rulli MC, Santini M, Hayman DTS, D’Odorico P. The nexus between forest fragmentation in Africa and Ebola virus disease outbreaks. Sci Rep. 2017;7(1):41613. Olivero J, et al. Recent loss of closed forests is associated with Ebola virus disease outbreaks. Sci Rep. 2017;7(1):14291. Morand S, Lajaunie C. Outbreaks of vector-borne and zoonotic diseases are associated with changes in forest cover and oil palm expansion at global scale. Front Vet Sci. 2021;8:661063. Vinson JE, et al. Land reversion and zoonotic spillover risk. R Soc open Sci. 2022;9(6):220582. Moyo E, Mhango M, Moyo P, Dzinamarira T, Chitungo I, Murewanhema G. Emerging infectious disease outbreaks in Sub-Saharan Africa: Learning from the past and present to be better prepared for future outbreaks. Front public Heal. 2023;11:1049986. Sachan N, Singh VP. Effect of climatic changes on the prevalence of zoonotic diseases. Vet World. 2010;3(11):519. Guimarães P, Keller A, Fehlmann T, Lammert F, Casper M. Deep-learning based detection of gastric precancerous conditions. Gut. 2020;69(1):4–6. Tabarestani S et al. Longitudinal Prediction Modeling of Alzheimer Disease using Recurrent Neural Networks, in., 2019 IEEE EMBS International Conference on Biomedical & Health Informatics (BHI) , 2019, pp. 1–4. 10.1109/BHI.2019.8834556 Masini RP, Medeiros MC, Mendes EF. Machine learning advances for time series forecasting. J Econ Surv. 2023;37(1):76–111. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6969004","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":521135305,"identity":"f9961c52-40d9-4d22-866d-a95b234a0fbd","order_by":0,"name":"Thomas James Wanyama","email":"data:image/png;base64,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","orcid":"","institution":"Makerere University","correspondingAuthor":true,"prefix":"","firstName":"Thomas","middleName":"James","lastName":"Wanyama","suffix":""}],"badges":[],"createdAt":"2025-06-24 21:53:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6969004/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6969004/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":92574505,"identity":"c7dddf8e-2b61-4ebf-a73a-ecdfaefd492f","added_by":"auto","created_at":"2025-10-01 08:12:43","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1232437,"visible":true,"origin":"","legend":"","description":"","filename":"RevisedManuscriptEbolaPrediction.docx","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/f19689dc27c5da72fcfcbc92.docx"},{"id":92574503,"identity":"8e305613-1afb-4fa1-925d-48193545bb7c","added_by":"auto","created_at":"2025-10-01 08:12:43","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4351,"visible":true,"origin":"","legend":"","description":"","filename":"92ab86e1b49f42bf9b6c2294b56ac340.json","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/388164c7e1a3e5fc7766a956.json"},{"id":92572556,"identity":"f36fed07-b249-49ab-a945-a143ed03be41","added_by":"auto","created_at":"2025-10-01 08:04:43","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":160746,"visible":true,"origin":"","legend":"","description":"","filename":"92ab86e1b49f42bf9b6c2294b56ac3401enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/f09274928fb86c7c3a1f11fa.xml"},{"id":92575314,"identity":"28bbbae5-3027-4192-98a3-18a5e49389ab","added_by":"auto","created_at":"2025-10-01 08:20:43","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1225,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/60d98566f60bf21d238360cc.png"},{"id":92572557,"identity":"69575d21-09f4-407d-9272-c13c431a0d5b","added_by":"auto","created_at":"2025-10-01 08:04:43","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":58878,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage10.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/008df9999a6d21d8124d6b6f.png"},{"id":92576639,"identity":"a76c2adc-5bff-441d-91b2-e503711fb462","added_by":"auto","created_at":"2025-10-01 08:28:44","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":71075,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage11.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/b3b4aac264e1c1c747778e89.png"},{"id":92575315,"identity":"2ae053ad-d6f6-4794-8a6e-3fb64df1bc55","added_by":"auto","created_at":"2025-10-01 08:20:43","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":67604,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage12.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/30595ddf57776c1381a04127.png"},{"id":92572566,"identity":"4a9d3ab2-acd5-4aad-88f0-a29929d77bb3","added_by":"auto","created_at":"2025-10-01 08:04:44","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":60249,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage13.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/09a2c9372e3e012fbc83bbdc.png"},{"id":92572572,"identity":"8c45f746-6948-4d39-a23f-9ea900c883ba","added_by":"auto","created_at":"2025-10-01 08:04:44","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":69789,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage14.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/aa54a855343a2b549e0e181d.png"},{"id":92572570,"identity":"b4d6d905-a8d1-4dd0-b2dd-7e2eddf450ff","added_by":"auto","created_at":"2025-10-01 08:04:44","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":66102,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage15.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/0c46834d6b564ceab35c0dae.png"},{"id":92574515,"identity":"105746c3-86c8-41c2-8f60-60a02a666eed","added_by":"auto","created_at":"2025-10-01 08:12:45","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":60077,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage16.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/47d3062bb63bf847e9b95c1f.png"},{"id":92572579,"identity":"bba9a241-306a-4159-a0b8-d06d20335c88","added_by":"auto","created_at":"2025-10-01 08:04:45","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":62958,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage17.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/aa34e2d235bc3476875b117c.png"},{"id":92574520,"identity":"a3d6786c-7385-400c-9d70-f770712b68d2","added_by":"auto","created_at":"2025-10-01 08:12:45","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":67713,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage18.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/0ce8c76bf13b1c019b6a01d6.png"},{"id":92572571,"identity":"65527ed6-3f5d-4d6e-8152-d19a3baaa7c2","added_by":"auto","created_at":"2025-10-01 08:04:44","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":67286,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage19.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/5d7400b4bcb1b090c303cacd.png"},{"id":92572596,"identity":"65ad779d-c0f7-410b-88fa-139d00faa324","added_by":"auto","created_at":"2025-10-01 08:04:45","extension":"jpeg","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":84285,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/900691eb5d63d7fbb2d3730b.jpeg"},{"id":92572601,"identity":"8eed675d-806b-450f-813b-74fbb23260aa","added_by":"auto","created_at":"2025-10-01 08:04:46","extension":"jpeg","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":32856,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/7d389fb3c060976fe244e283.jpeg"},{"id":92575325,"identity":"177dc73c-0562-45f9-af1a-b6cff85d8083","added_by":"auto","created_at":"2025-10-01 08:20:45","extension":"jpeg","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":34854,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/a5a14367256ffe21e86069e4.jpeg"},{"id":92574525,"identity":"9e80cda3-8388-489e-8ee7-3d352153ef68","added_by":"auto","created_at":"2025-10-01 08:12:46","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":69936,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/68d8ab79dc71a032bf4167e5.png"},{"id":92572605,"identity":"8d588f7a-0dd0-4ed6-8073-5a5587f91de4","added_by":"auto","created_at":"2025-10-01 08:04:46","extension":"jpeg","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":98361,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/a4b201f21fafdd2a7aaefffa.jpeg"},{"id":92572607,"identity":"10a770c9-c895-4e02-b5f2-6caba2e7e2db","added_by":"auto","created_at":"2025-10-01 08:04:46","extension":"jpeg","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":30365,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage7.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/5cd7fe88469712cf10a1c5e9.jpeg"},{"id":92572586,"identity":"4e5ee09b-7dd9-4882-b171-2628b7861557","added_by":"auto","created_at":"2025-10-01 08:04:45","extension":"png","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":68708,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/c58c77c01404df69201e4da7.png"},{"id":92572576,"identity":"bbfddfef-6e64-46ad-8025-82f17fe1232e","added_by":"auto","created_at":"2025-10-01 08:04:45","extension":"png","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":66012,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/647b4de26b35c1326aa9f7be.png"},{"id":92572604,"identity":"abbbec9a-45a5-4607-ae71-7a28fe18eaef","added_by":"auto","created_at":"2025-10-01 08:04:46","extension":"png","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":851,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/48d01d4651e45df278630481.png"},{"id":92574523,"identity":"6a16d548-b299-4d4a-9470-d7c7a6c41c6e","added_by":"auto","created_at":"2025-10-01 08:12:45","extension":"png","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":18855,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage10.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/4324c316b5d24d68f4db0838.png"},{"id":92574526,"identity":"7957d321-d422-402a-8f87-ce3d754f5ef9","added_by":"auto","created_at":"2025-10-01 08:12:46","extension":"png","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":22726,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage11.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/93ed5a60c1fd8e5d008a2369.png"},{"id":92572592,"identity":"952a0e57-f220-40d6-8391-ac67ed5e1cfd","added_by":"auto","created_at":"2025-10-01 08:04:45","extension":"png","order_by":25,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":23093,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage12.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/41c4b721071d4287a8d45881.png"},{"id":92572590,"identity":"20b93aa2-d856-4612-a584-b75dcc397148","added_by":"auto","created_at":"2025-10-01 08:04:45","extension":"png","order_by":26,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":19382,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage13.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/cafc4bd0f2812674664b08be.png"},{"id":92572606,"identity":"a67fb85f-7dac-46e2-821c-958d573970dd","added_by":"auto","created_at":"2025-10-01 08:04:46","extension":"png","order_by":27,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":22561,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage14.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/6ec0f28240a7cfbb469e7a08.png"},{"id":92572578,"identity":"2085ba8a-1469-4b31-9170-77ad493fa362","added_by":"auto","created_at":"2025-10-01 08:04:45","extension":"png","order_by":28,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":22285,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage15.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/1df66f10ab6a5c8af4bb8f80.png"},{"id":92572591,"identity":"9c226000-17de-43d3-80e5-ded83c769cb5","added_by":"auto","created_at":"2025-10-01 08:04:45","extension":"png","order_by":29,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":19391,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage16.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/6e7f9f39453915c0116f2a32.png"},{"id":92572575,"identity":"896d6bd2-cc92-4bae-bef9-eb6e489607b7","added_by":"auto","created_at":"2025-10-01 08:04:45","extension":"png","order_by":30,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":23243,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage17.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/559892f210d0f685a6921151.png"},{"id":92572589,"identity":"ff8c8c54-77d1-4100-9ccf-b2f582a33f82","added_by":"auto","created_at":"2025-10-01 08:04:45","extension":"png","order_by":31,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":22989,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage18.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/9eafcc544a0eb5eb9e91eb9c.png"},{"id":92575330,"identity":"0f188ea3-7a85-4f53-a5a3-c85ea654bc6d","added_by":"auto","created_at":"2025-10-01 08:20:46","extension":"png","order_by":32,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":21265,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage19.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/456b3df1dcb47fead451cd6e.png"},{"id":92572599,"identity":"bdcfada9-7ab1-43dd-b47e-e9079cc3d751","added_by":"auto","created_at":"2025-10-01 08:04:46","extension":"png","order_by":33,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":58026,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/aa075c9442d5328f26d61990.png"},{"id":92572600,"identity":"0ffe4d6e-0e24-4d09-b6b2-d229cf6c0145","added_by":"auto","created_at":"2025-10-01 08:04:46","extension":"png","order_by":34,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":11242,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/d50d79a06744d0fc84e7da2b.png"},{"id":92572598,"identity":"b6c5c248-ce14-406e-84b3-0bebba1d8cf6","added_by":"auto","created_at":"2025-10-01 08:04:45","extension":"png","order_by":35,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":27114,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/3962dd47b1c49831392ba1d4.png"},{"id":92572603,"identity":"4ff43c5e-d5f6-433d-a9ec-4b98a01f92af","added_by":"auto","created_at":"2025-10-01 08:04:46","extension":"png","order_by":36,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":63931,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/504897b53fa9862dea6d47be.png"},{"id":92574529,"identity":"eff15d3d-8816-4de9-b548-ac951d3acf9f","added_by":"auto","created_at":"2025-10-01 08:12:46","extension":"png","order_by":37,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":17734,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/a90cdb1be21402f411d68c66.png"},{"id":92575328,"identity":"43263508-12ba-497e-8c07-4fcd1015360b","added_by":"auto","created_at":"2025-10-01 08:20:45","extension":"png","order_by":38,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":22063,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/7e1544c5e737a71d2414781d.png"},{"id":92572609,"identity":"7012f2fa-6aca-4d65-81a8-faa2546cb69e","added_by":"auto","created_at":"2025-10-01 08:04:46","extension":"png","order_by":39,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":22058,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/6d0291dfceba2f3e259b72bd.png"},{"id":92572594,"identity":"060d7686-8c26-425f-b4b2-c8c02ca6b7fa","added_by":"auto","created_at":"2025-10-01 08:04:45","extension":"png","order_by":40,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":22054,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/558d214bad53983c022d70d6.png"},{"id":92572610,"identity":"1c4ba3df-68e9-4f5f-a225-12abb390b2e8","added_by":"auto","created_at":"2025-10-01 08:04:46","extension":"xml","order_by":41,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":159319,"visible":true,"origin":"","legend":"","description":"","filename":"92ab86e1b49f42bf9b6c2294b56ac3401structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/37cb336ff55947458b2b31a2.xml"},{"id":92575332,"identity":"7b4520a7-6886-477a-b310-97d22c7dd481","added_by":"auto","created_at":"2025-10-01 08:20:46","extension":"html","order_by":42,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":173593,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/775ea2dcfb0da11e03f86cd2.html"},{"id":92572554,"identity":"ab069cea-e0ac-4f60-9f23-052ed6ecc2ad","added_by":"auto","created_at":"2025-10-01 08:04:43","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":53714,"visible":true,"origin":"","legend":"\u003cp\u003eFlow chart of the methodology\u003c/p\u003e","description":"","filename":"Picture1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/4fd95d4b931fde94c1ec2790.jpg"},{"id":92572552,"identity":"2a30f4c7-30d7-476a-9495-794b1365beb6","added_by":"auto","created_at":"2025-10-01 08:04:43","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":41003,"visible":true,"origin":"","legend":"\u003cp\u003eSimple architecture of the gradient boosting algorithm [27]\u003c/p\u003e","description":"","filename":"Picture2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/ddfd57a6db771e60574d4f54.jpg"},{"id":92572553,"identity":"07a25b16-39c6-42ec-9a58-4526f354d629","added_by":"auto","created_at":"2025-10-01 08:04:43","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":73410,"visible":true,"origin":"","legend":"\u003cp\u003eSimple architecture of the decision tree algorithm [33]\u003c/p\u003e","description":"","filename":"Picture3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/795ced6e0d7032f5b13cea27.jpg"},{"id":92574507,"identity":"a63aa304-96f0-4494-b3cf-5cf2af73a4cf","added_by":"auto","created_at":"2025-10-01 08:12:43","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":92570,"visible":true,"origin":"","legend":"\u003cp\u003eSimple architecture of a deep neural network [35]\u003c/p\u003e","description":"","filename":"Picture4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/e982a25b8a706b9be797a7be.jpg"},{"id":92577028,"identity":"33809e4b-7df9-4ede-9c04-fde9ac4c7a09","added_by":"auto","created_at":"2025-10-01 08:36:43","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":33162,"visible":true,"origin":"","legend":"\u003cp\u003eSimple Architecture of an artificial neural network [39]\u003c/p\u003e","description":"","filename":"Picture5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/242dae1086d82f5c6505c4e6.jpg"},{"id":92572560,"identity":"ace587b3-c06b-43b3-852c-6ca8df495423","added_by":"auto","created_at":"2025-10-01 08:04:43","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":43626,"visible":true,"origin":"","legend":"\u003cp\u003eMetrics used to evaluate machine learning algorithms\u003c/p\u003e","description":"","filename":"Picture6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/19d9c76287f4a1970392f358.jpg"},{"id":92574510,"identity":"7f35b0ec-1352-4197-9148-2453a9b6c222","added_by":"auto","created_at":"2025-10-01 08:12:43","extension":"jpg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":58031,"visible":true,"origin":"","legend":"\u003cp\u003eHistorical and predicted Ebola cases and deaths in Guinea\u003c/p\u003e","description":"","filename":"Picture7.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/e9d875b2c79891a909a4c8f3.jpg"},{"id":92574512,"identity":"6555b11f-c5b6-44c0-a67b-6e36485d73eb","added_by":"auto","created_at":"2025-10-01 08:12:44","extension":"jpg","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":63405,"visible":true,"origin":"","legend":"\u003cp\u003eHistorical and predicted Ebola cases and deaths in Liberia\u003c/p\u003e","description":"","filename":"Picture8.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/780976341970c453059476e8.jpg"},{"id":92575323,"identity":"92b6c672-a882-4eba-b197-b887299f44d7","added_by":"auto","created_at":"2025-10-01 08:20:45","extension":"jpg","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":56083,"visible":true,"origin":"","legend":"\u003cp\u003eHistorical and predicted Ebola cases and deaths in Sierra Leone\u003c/p\u003e","description":"","filename":"Picture9.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/f5939ad594e2ca1b1a9caafa.jpg"},{"id":92575322,"identity":"6a6e95dd-4c8c-4f8b-9135-79b39cc4e831","added_by":"auto","created_at":"2025-10-01 08:20:44","extension":"jpg","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":60821,"visible":true,"origin":"","legend":"\u003cp\u003eHistorical and predicted Ebola cases and deaths in Guinea\u003c/p\u003e","description":"","filename":"Picture10.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/0b4239161b0a0a0370fb7df3.jpg"},{"id":92576643,"identity":"304ceebc-444e-43ea-b852-8f4d8da20c2f","added_by":"auto","created_at":"2025-10-01 08:28:45","extension":"jpg","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":69012,"visible":true,"origin":"","legend":"\u003cp\u003eHistorical and predicted Ebola cases and deaths in Liberia\u003c/p\u003e","description":"","filename":"Picture11.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/4af6d7d27ab0b90fb8c9c15d.jpg"},{"id":92572568,"identity":"201b1cd4-e11f-40ce-bcb9-8f29c0c0fe18","added_by":"auto","created_at":"2025-10-01 08:04:44","extension":"jpg","order_by":12,"title":"Figure 12","display":"","copyAsset":false,"role":"figure","size":58533,"visible":true,"origin":"","legend":"\u003cp\u003eHistorical and predicted Ebola cases and deaths in Sierra Leone\u003c/p\u003e","description":"","filename":"Picture12.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/b609dfb9420a791f86130bb7.jpg"},{"id":92574511,"identity":"17bdda70-8e66-4ecb-8ee8-c4040a35a8e0","added_by":"auto","created_at":"2025-10-01 08:12:44","extension":"jpg","order_by":13,"title":"Figure 13","display":"","copyAsset":false,"role":"figure","size":58938,"visible":true,"origin":"","legend":"\u003cp\u003eHistorical and predicted Ebola cases and deaths in Guinea\u003c/p\u003e","description":"","filename":"Picture13.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/898e338cb179e0201c5a3a63.jpg"},{"id":92574519,"identity":"137ee5e9-3e81-4899-973a-d9405de0517d","added_by":"auto","created_at":"2025-10-01 08:12:45","extension":"jpg","order_by":14,"title":"Figure 14","display":"","copyAsset":false,"role":"figure","size":56396,"visible":true,"origin":"","legend":"\u003cp\u003eHistorical and predicted Ebola cases and deaths in Liberia\u003c/p\u003e","description":"","filename":"Picture14.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/b50bae0641cc706b97872389.jpg"},{"id":92572597,"identity":"c5052ce8-1bd4-46d1-a706-950f703f278e","added_by":"auto","created_at":"2025-10-01 08:04:45","extension":"jpg","order_by":15,"title":"Figure 15","display":"","copyAsset":false,"role":"figure","size":51972,"visible":true,"origin":"","legend":"\u003cp\u003eHistorical and predicted Ebola cases and deaths in Sierra Leone\u003c/p\u003e","description":"","filename":"Picture15.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/84c3b8f1997e367c96410203.jpg"},{"id":92576642,"identity":"89974504-433a-49c8-9a1a-4e733a41b4cf","added_by":"auto","created_at":"2025-10-01 08:28:45","extension":"jpg","order_by":16,"title":"Figure 16","display":"","copyAsset":false,"role":"figure","size":56984,"visible":true,"origin":"","legend":"\u003cp\u003eHistorical and predicted Ebola cases and deaths in Guinea\u003c/p\u003e","description":"","filename":"Picture16.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/e8138471719b2e35641c91c0.jpg"},{"id":92574514,"identity":"68a1751c-4df3-484d-822d-7a6f4254f02b","added_by":"auto","created_at":"2025-10-01 08:12:44","extension":"jpg","order_by":17,"title":"Figure 17","display":"","copyAsset":false,"role":"figure","size":61885,"visible":true,"origin":"","legend":"\u003cp\u003eHistorical and predicted Ebola cases and deaths in Liberia\u003c/p\u003e","description":"","filename":"Picture17.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/7060ae081957c4cf26fc0e76.jpg"},{"id":92572595,"identity":"7f2ae1a6-c3e0-4197-b8b8-267f3ef85f28","added_by":"auto","created_at":"2025-10-01 08:04:45","extension":"jpg","order_by":18,"title":"Figure 18","display":"","copyAsset":false,"role":"figure","size":61293,"visible":true,"origin":"","legend":"\u003cp\u003eHistorical and predicted Ebola cases and deaths in Sierra Leone\u003c/p\u003e","description":"","filename":"Picture18.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/ad7ef385d3832d3ba7e91dac.jpg"},{"id":92600385,"identity":"e263026c-2ea3-445b-854a-327ea09f137d","added_by":"auto","created_at":"2025-10-01 14:21:42","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2682135,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6969004/v1/5622777e-0b3b-4ff9-956f-c2ccf6ac5315.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Data-driven Modeling as a Tool for Prediction of Future Outbreaks of Ebola Virus Disease in West Africa","fulltext":[{"header":"1. Background","content":"\u003cp\u003eEbola virus disease (EVD) is a complex zoonosis that has been reported as being highly virulent in humans [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. It is among the deadliest viral diseases in human history due to its high death rates [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. EVD is initiated by the Ebola virus (EBOV) [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Some of the symptoms of EVD are: initial fever, fatigue before descending into headaches, vomiting, violent diarrhea, multiple organ failure, and massive internal bleeding [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Globally, one of the regions that is affected by EVD is Sub-Saharan Africa, since it has continually battled this fatal disease with frequent outbreaks [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. For example, the 2014 Ebola outbreak is reported to have claimed 11306 deaths out of 28200 cases in 10 West African countries [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Some of these countries include: Guinea, Liberia, Mali, Nigeria, Senegal, and Sierra Leone. This 2014 Ebola outbreak held a greater magnitude than all the other previous outbreaks [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eThis West African outbreak had never been reported previously in the region despite Ebola being known to cause outbreaks in central and eastern Africa [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Some of the reasons behind the prevalent nature of the West African outbreak include the highly mobility of the West African communities and their gigantic population [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Other factors include climate change [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e], urbanization, and the rising demand for food that is got from animals [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. There have been efforts to handle any EVD outbreak that may occur up in the near future [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. However, these efforts may not be fully effective as they may present some back and forth challenges. Therefore, relying on smart measures such as data-driven modeling is imperative. This data-driven modeling encompasses the use of forecasting models which utilize maximum data for reducing bias and permit automatic selection of predicting variables [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eIt can serve as an effective early warning tool in disease prediction. For example, data-driven modeling approaches such as machine learning have been used as effective tools to predict the mortality risk of COVID-19 patients [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. Another study reported the use of tree-based machine learning (ML) algorithms and feed-forward neural networks as effective tools for early predictions of Kyasanur forest disease, an emerging zoonotic disease [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. Statistical models such as logistic regression and ML models such as support vector machines, random forests, and decision trees have been used in the prediction of COVID-19 [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. Time-series deep learning models such as the improved Long Short-Term Memory deep learning (DL) method have also been used in Iran, Russia, and Peru to predict the epidemic trends of COVID-19 [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eOther than the improved Long Short-Term Memory DL method, other Time-series DL models such as Recurrent Neural Network (RNN) and Back Propagation Neural Network (BPNN) have been used in the prediction of Hepatitis B in China [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. In the present study, regression-based models, DL models, and a hybrid model were utilized to predict EVD cases and deaths in Guinea, Liberia, and Sierra Leone for the next 10 years. To the best of the author\u0026rsquo;s knowledge, this is the first report that has used DT, GB, DNN, and a hybrid model (built from ANN, DT, and GB) to predict an Ebola outbreak in all the three mentioned countries.\u003c/p\u003e\u003cp\u003eThe three countries were selected because of the high Ebola cases and deaths recorded compared to other countries in West Africa (Nigeria, Senegal, and Mali), where Ebola cases and deaths were minimal. Despite immense studies on Ebola outbreak in the three countries [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e], [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e], [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e], [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e], [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e], [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e] and with other studies covering the disease\u0026rsquo;s forecasting, none of these studies have considered forecasting of both Ebola cases and deaths using multiple models involving tree-based ML models, DL models, and a hybrid model. Yet having a comprehensive study involving multiple forecasting models is key since different models capture patterns differently. Regression-based ML algorithms were chosen over time-series forecasting approaches due to their ability to process non-linear and complex data, which is a big limitation for time series models [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. Regression algorithms are also comparatively insensitive to co-variation among the input predictors and are non-parametric thus offering them an advantage over time series models [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e].\u003c/p\u003e"},{"header":"2. Methods","content":"\u003cp\u003eThis section presents information on the dataset, and the steps done to prepare the data for forecasting. A summary of this section is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1 Data collection\u003c/h2\u003e\u003cp\u003eThe dataset was obtained from an online repository accessible at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.kaggle.com/datasets/imdevskp/ebola-outbreak-20142016-complete-dataset\u003c/span\u003e\u003cspan address=\"https://www.kaggle.com/datasets/imdevskp/ebola-outbreak-20142016-complete-dataset\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. This dataset contained data on the EVD outbreak in West Africa recorded from August 2014 to March 2016 (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eDataset and its features\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"3\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eS/n\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eDataset Name\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eFeatures in the dataset\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\" morerows=\"3\" rowspan=\"4\"\u003e\u003cp\u003eEbola | 2014\u0026ndash;2016 | Western Africa Ebola Outbreak\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eCountry\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eYear of reporting\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eCumulative number of Ebola cases (confirmed, probable and suspected cases)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eCumulative number of Ebola deaths\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2 Data pre-processing\u003c/h2\u003e\u003cp\u003eThis step involved filtering and modifying the data so that it is easy to explore and understand. For example, the dataset that was obtained from the online repository contained all the data for the three countries. However, for easy exploration, data for each country were obtained from the overall dataset and then saved as an independent file. Additionally, the columns on cumulative number of confirmed, probable, and suspected cases and deaths were renamed as Ebola cases and deaths, respectively. Data normalization was also carried out.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e2.3 Feature engineering\u003c/h2\u003e\u003cp\u003eA feature is a characteristic of any object. In machine learning, a sample is disintegrated into a set of features before training and testing for tasks, such as classification, prediction, or clustering [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. In the current study, feature engineering was used to extract relevant features (data on year, Ebola cases, and deaths) from the dataset.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003e2.4 Learning algorithms\u003c/h2\u003e\u003cp\u003eThe algorithms used were decision trees (DT), gradient boosting (GB), deep neural networks (DNN), and a hybrid model (built from decision trees, gradient boosting, and artificial neural networks). A brief description of these algorithms is given below.\u003c/p\u003e\u003cp\u003e\u003cb\u003eGradient boosting\u003c/b\u003e; this is a powerful and efficient learning method that is based on an implementation of gradient-boosted decision trees, and it is mostly deployed for supervised learning tasks, mainly regression and classification problems [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. As an iterative approach, gradient boosting creates numerous trees in a sequential order of successive decision trees [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. It has been complemented with a feature selection process that increases the overall performance by extracting the most relevant features from the input data [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e].\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eDecision trees\u003c/b\u003e; Decision trees have been utilized in many applications such as classification, regression, and feature selection [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. The idea behind decision trees is that they recursively divide the data into subsets based on the values of different attributes until a stopping criterion is met [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. The end-result is a tree-like structure, where each node represents a decision or a split based on a specific attribute [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. This tree structure makes it possible for users to understand and interpret the decision-making process easily [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. While using decision trees, a sequence of Boolean decisions are made with the aim to group data into categorical bins [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. Decision trees also represent a mapping relationship between object attributes and object values, and their basic process follows the \"divide and conquer\" strategy [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e].\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eDeep neural networks\u003c/b\u003e; deep neural networks are part of the neural networks that are built to mimic the biological structures of the human brain [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. They work best in the over-parameterized regime, with many more parameters than data points [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. They are composed of units called neurons, which are connected to each other in a network [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e].\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eArtificial neural networks\u003c/b\u003e; these were developed to copy the features of the biological neurons in the human brain and the nervous system, and they are able to keep the biological concept of artificial neurons [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. They consist of initial input data, activation function, and producing output with an output function, while the activation function can provide a smooth transition as input values modify [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. The ANN is composed of connections, while each connection is indicated by a weight as its related importance, which can provide the output of one neuron as the input of another neuron [\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]. In the Ebola forecasting modeling, the historical incidence is used as the input neurons, while the related predicting incidence is obtained from the output neurons after the ANN is well trained.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\u003ch2\u003e2.5 Computation of feature importance\u003c/h2\u003e\u003cp\u003eAs a key step in data-driven modeling, feature importance (FI) helps to provide an understanding of how decisions are made and how this enhances the establishment of true causality between vital data attributes and outcomes in model inference [\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e]. In this study, FI scores were calculated to reflect the input features for the four models [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e] since different models are known to produce different FI values due to variations in their learning algorithms [\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]. Year and cases were taken as the input features while the target was death.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003e2.6 Model hyper-parameter tuning\u003c/h2\u003e\u003cp\u003ePrior to training the data, hyper-parameters were carefully configured so as to optimize the performance of the different algorithms. This process particularly involved adjusting the number of trees, learning cycles, learning rate, size of layers, and the number of training epochs.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\u003ch2\u003e2.7 Training of algorithms\u003c/h2\u003e\u003cp\u003eTraining of the algorithms was done using Matlab R2024a software. It involved splitting the data into the training and testing sets were done. 80% of the data were incorporated into the training set, and 20% added to the testing set.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003e2.8 Data Visualization\u003c/h2\u003e\u003cp\u003eGraphical representations of the predicted data on EVD cases and deaths were performed on Matlab R2024a (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.mathworks.com/\u003c/span\u003e\u003cspan address=\"http://www.mathworks.com/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), and this helped to identify trends and patterns in the EVD outbreak.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003e2.9 Evaluation Criteria\u003c/h2\u003e\u003cp\u003eEvaluation of the algorithms was done using three performance metrics. These were; mean absolute scaled error (MASE), root mean squared error (RMSE), and coefficient of determination (R\u0026sup2;) (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e). These statistics helped to compare the target and output values, and also calculate a score as an index for the performance and accuracy of the forecasting algorithms [\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e].\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003e2.10 Uncertainty Quantification\u003c/h2\u003e\u003cp\u003eThe processes of ML and DL have diverse sources of uncertainty such as vagueness due to extrapolation, variance in model parameters, inherent noise in data, and appropriateness of model selection [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e]. Quantifying these underlying uncertainties is crucial as it helps to establish trust, determine risk in alternatives, or communicate the potential for error [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e]. In the present study, uncertainty quantification (UQ) was done using the bootstrapping method to explore the confidence in the predictions by decision trees, gradient boosting, deep neural networks, and the hybrid model.\u003c/p\u003e\u003c/div\u003e"},{"header":"3. Results","content":"\u003cp\u003eThis section presents the historical data on Ebola cases and deaths from 2013 to 2016 in Guinea, Liberia, and Sierra Leone. It goes ahead to present the likely number of EVD cases and deaths in the next ten years from when the EVD outbreak was last recorded in the three West African countries.\u003c/p\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003e3.1 Historical Ebola cases and deaths\u003c/h2\u003e\u003cp\u003eAccording to the year, the mean Ebola cases and deaths were extremely high in 2015 (i.e. 12,714 cases and 3,786 deaths), nearly twice the next highest mean cases in 2014 (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The mean EVD cases and deaths from 2013 to 2016 by country were high in Sierra Leone (i.e. 11,596 cases and 3,432 deaths) compared to other countries (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eIn Guinea, the historical Ebola cases and deaths were extremely high in 2014 and then underwent a constant increase from 2014 to 2015, followed by a sharp increase from 2015 to 2016. In Liberia, the historical Ebola cases and deaths were immensely high in 2014 and then underwent a constant increase from 2014 to 2015, followed by a sharp increase from 2015 to 2016. In Sierra Leone, the historical Ebola cases and deaths were tremendously high in 2014 and then underwent a constant increase from 2014 to 2015, followed by a sharp increase from 2015 to 2016.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eThe historical EVD cases and deaths by country and year, including the mean, minimum, and maximum of the cases and deaths\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e\u003cp\u003eHistorical Ebola cases\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e\u003cp\u003eHistorical Ebola deaths\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCountry\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003eYear\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003eMean\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003eMin-Max\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003eMean\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003eMin-Max\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e\u003cp\u003eGuinea\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2014\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1741.842\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e648\u0026ndash;2707\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1064.789\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e430\u0026ndash;1709\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2015\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3564.200\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2730\u0026ndash;3810\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e2363.959\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1739\u0026ndash;2536\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2016\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e\u003cp\u003eLiberia\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2014\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e5659.946\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1871\u0026ndash;8018\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e2565.703\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1089\u0026ndash;3423\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2015\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e6906.315\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e4-10666\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e3084.552\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1-4806\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2016\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3560\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e5-10666\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1604.333\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e3-4806\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e\u003cp\u003eSierra Leone\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2014\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e5060.289\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1026\u0026ndash;9446\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1366.974\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e422\u0026ndash;2758\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2015\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e12713.7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e9633\u0026ndash;14122\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e3786.064\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e2827\u0026ndash;3955\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2016\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eSummary of the historical case and death counts for Ebola Viral Disease by country, including the mean, minimum, and maximum of the cases and deaths.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e\u003cp\u003eHistorical Ebola cases\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003eHistorical Ebola deaths\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCountry\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003eMean\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003eMin-Max\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003eMean\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003eMin-Max\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGuinea\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3297.75\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e648\u0026ndash;3810\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2174.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e430\u0026ndash;2536\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLiberia\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e6752.04\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4-10666\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e3019.61\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1-4806\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSierra Leone\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e11596.25\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1026\u0026ndash;14122\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e3431.79\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e422\u0026ndash;3955\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003e3.2 Feature importance results\u003c/h2\u003e\u003cp\u003eOverall, cases as a feature input had higher FI scores than year for the four models, implying that variations in Ebola cases are strongly predictive of Ebola deaths in Guinea, Liberia, and Sierra Leone (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). The higher FI scores for the year feature in Guinea using the gradient boosting model reveal strong time-based patterns, such as increasing or reducing Ebola deaths over time in Guinea than in other countries. Overall, the decision tree model had higher FI scores for cases in two countries compared to other models (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eFI scores for the different models\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCountry\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eInput features\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eDT\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eGB\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eDNN\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eHybrid model\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eGuinea\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eYear\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e8.94934E-06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.000473691\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.000534377\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eCases\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.999991051\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.999526322\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.999465623\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eLiberia\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eYear\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e3.38267E-05\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e4.89429E-06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.9976e-05\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eCases\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.999966173\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.999995112\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.999980024\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eSierra Leone\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eYear\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.001029601\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.001291345\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.000134385\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.000781894\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eCases\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.998970399\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.998708655\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.999865592\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.999218106\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003e3.3 Visualization of EVD case and death forecasts in the three countries\u003c/h2\u003e\u003cp\u003e\u003cb\u003eVisualization of EVD case and death forecasts from decision trees\u003c/b\u003e\u003c/p\u003e\u003cp\u003ePrediction using DT revealed a constant increase in Ebola cases and deaths in Guinea, Liberia, and Sierra Leone from 2017 to 2026 (Figs.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e, \u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e, \u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e). The predicted Ebola case counts were higher than the Ebola death counts in the three countries. Among the three countries, higher Ebola case and death counts (about 14000 cases and 4000 deaths) were registered in Sierra Leone. Guinea followed second with higher case and death counts (about 3800 cases and 2500 deaths) than Liberia.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eVisualization of EVD case and death forecasts from gradient boosting\u003c/b\u003e\u003c/p\u003e\u003cp\u003eForecasting using GB showed a constant increase in Ebola cases and deaths in all the countries from 2017 to 2026 (Figs.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003e, \u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003e, \u003cspan refid=\"Fig12\" class=\"InternalRef\"\u003e12\u003c/span\u003e). Additionally, the predicted Ebola case counts were higher than the Ebola death counts in the three countries. Of all the three countries, higher Ebola case and death counts (about 14000 cases and 4000 deaths) were recorded in Sierra Leone. Liberia followed second with higher case counts (about 7000 cases) than Guinea. However, the death counts in Guinea and Sierra Leone were similar (about 3000 deaths).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eVisualization of EVD case and death forecasts from deep neural networks\u003c/b\u003e\u003c/p\u003e\u003cp\u003ePrediction using DNN revealed a continual increase in Ebola cases and deaths in all the countries from 2017 to 2026 (Figs.\u0026nbsp;\u003cspan refid=\"Fig13\" class=\"InternalRef\"\u003e13\u003c/span\u003e, \u003cspan refid=\"Fig14\" class=\"InternalRef\"\u003e14\u003c/span\u003e, \u003cspan refid=\"Fig15\" class=\"InternalRef\"\u003e15\u003c/span\u003e). The projected Ebola case counts were higher than the Ebola death counts in the three countries. Extremely high Ebola case and death counts (about 14000 cases and 3900 deaths) were recorded in Sierra Leone. This was followed by Liberia (approximately 4000 cases and 3800 deaths), and then Guinea (approximately 3750 cases and 2500 deaths).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eVisualization of EVD case and death forecasts from the hybrid model\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe hybrid model revealed a sharp increase in the Ebola cases and deaths in Guinea and Sierra Leone from 2017 to 2026 (Figs.\u0026nbsp;\u003cspan refid=\"Fig16\" class=\"InternalRef\"\u003e16\u003c/span\u003e and \u003cspan refid=\"Fig18\" class=\"InternalRef\"\u003e18\u003c/span\u003e). However, in Liberia, Ebola cases were anticipated to gradually increase while the Ebola death counts were shown to gradually decrease from 2017 to 2026 (Fig.\u0026nbsp;\u003cspan refid=\"Fig17\" class=\"InternalRef\"\u003e17\u003c/span\u003e). The prediction also revealed higher Ebola case and death counts in Sierra Leone than in other countries.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\u003ch2\u003e3.4 Performance of the forecasting models based on the country\u003c/h2\u003e\u003cp\u003eDecision trees, gradient boosting, deep neural networks, and the hybrid model were used as the forecasting algorithms to predict the next Ebola cases and deaths in Guinea, Liberia, and Sierra Leone.\u003c/p\u003e\u003cp\u003e\u003cb\u003ePerformance of the decision tree model\u003c/b\u003e\u003c/p\u003e\u003cp\u003eBased on the country, the decision tree model performed well by R\u003csup\u003e2\u003c/sup\u003e and MASE in predicting Ebola cases and deaths in Liberia (Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). By RMSE, the model had a higher accuracy in predicting Ebola cases and deaths in Guinea than in other countries.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003ePerformance metrics of the decision trees by country\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eCases\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eDeaths\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCountry\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003eMASE\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003eRMSE\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003eR\u003c/b\u003e\u003csup\u003e\u003cb\u003e2\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003eMASE\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003eRMSE\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003eR\u003c/b\u003e\u003csup\u003e\u003cb\u003e2\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGuinea\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e19.75\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e380.96\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.74\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e25.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e258.88\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.76\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLiberia\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e0.89\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4590.79\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.87\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e2050.45\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.01\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSierra Leone\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e22.11\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1566.52\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e0.75\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e16.18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e388.17\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e0.83\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003ePerformance of gradient boosting model\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThis model had a good accuracy and good performance by RMSE and R\u003csup\u003e2\u003c/sup\u003e respectively in predicting Ebola cases and deaths in Guinea than in other countries (Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). By MASE, the model performed poorly in predicting Ebola cases and deaths than the na\u0026iuml;ve forecasts in all the three countries.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab6\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003ePerformance metrics of gradient boosting by country\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eCases\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eDeaths\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCountry\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003eMASE\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003eRMSE\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003eR\u003c/b\u003e\u003csup\u003e\u003cb\u003e2\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003eMASE\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003eRMSE\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003eR\u003c/b\u003e\u003csup\u003e\u003cb\u003e2\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGuinea\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e4.57\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e377.66\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e0.61\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e4.83\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e260.92\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e0.83\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLiberia\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4715.79\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1.11\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e2108.44\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.01\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSierra Leone\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e5.26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1873.38\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.53\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e4.86\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e505.03\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.61\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003ePerformance of deep neural networks\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe DNN model had the best accuracy by RMSE in predicting Ebola cases and deaths in Guinea than in other countries while a good performance in the prediction by R\u003csup\u003e2\u003c/sup\u003e was observed in Sierra Leone (Table\u0026nbsp;\u003cspan refid=\"Tab7\" class=\"InternalRef\"\u003e7\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab7\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 7\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003ePerformance metrics of deep neural network by country\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eCases\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eDeaths\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCountry\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003eMASE\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003eRMSE\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003eR\u003c/b\u003e\u003csup\u003e\u003cb\u003e2\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003eMASE\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003eRMSE\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003eR\u003c/b\u003e\u003csup\u003e\u003cb\u003e2\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGuinea\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eInf\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e385.98\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.73\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eInf\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e271.21\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.74\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLiberia\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eInf\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4610.21\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eInf\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e2059.30\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSierra Leone\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eInf\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1590.76\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e0.74\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eInf\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e401.80\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e0.82\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003ePerformance of the hybrid model\u003c/b\u003e\u003c/p\u003e\u003cp\u003eBy R\u003csup\u003e2\u003c/sup\u003e and MASE, the hybrid model performed better in predicting Ebola cases and deaths in Liberia than in other countries (Table\u0026nbsp;\u003cspan refid=\"Tab8\" class=\"InternalRef\"\u003e8\u003c/span\u003e). A good accuracy by RMSE in predicting Ebola cases and deaths was also registered in Guinea than other countries.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab8\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 8\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003ePerformance metrics of the hybrid model by country\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eCases\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eDeaths\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCountry\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003eMASE\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003eRMSE\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003eR\u003c/b\u003e\u003csup\u003e\u003cb\u003e2\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003eMASE\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003eRMSE\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003eR\u003c/b\u003e\u003csup\u003e\u003cb\u003e2\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGuinea\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e73.09\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e1143.32\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-1.33\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e89.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e761.38\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e-1.08\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLiberia\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e0.77\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3826.11\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e0.31\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.76\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1695.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e0.32\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSierra Leone\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e70.40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4085.15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-0.70\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e67.40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1208.70\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e-0.65\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e\u003ch2\u003e3.5 Overall model performance\u003c/h2\u003e\u003cp\u003eAll models performed poorly in predicting Ebola cases and deaths compared to the na\u0026iuml;ve forecast. By RMSE and R\u003csup\u003e2\u003c/sup\u003e, the decision tree model had a higher accuracy and performance better in predicting Ebola cases and deaths than other models (Table\u0026nbsp;\u003cspan refid=\"Tab9\" class=\"InternalRef\"\u003e9\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab9\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 9\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003ePerformance of the forecasting models\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eMASE\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eRMSE\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eR\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eMASE\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eRMSE\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eR\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eCases\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eDeaths\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDecision trees\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e14.25\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e2179.42\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e0.50\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e14.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e899.17\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e0.53\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGradient boosting\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3.65\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2322.28\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.38\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e3.60\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e958.13\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.48\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDeep neural networks\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eInf\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2195.65\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.49\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eInf\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e910.77\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.52\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHybrid model\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e48.09\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3018.20\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e-0.57\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e52.39\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1221.70\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e-0.47\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e\u003ch2\u003e3.6 Uncertainty quantification results by model\u003c/h2\u003e\u003cp\u003e\u003cb\u003eDecision tree\u003c/b\u003e; there was a greater uncertainty in the prediction of Ebola cases in 2017 and 2018 by decision trees. There was a high confidence in the prediction of Ebola cases from 2019 to 2026 by decision trees (Table\u0026nbsp;\u003cspan refid=\"Tab10\" class=\"InternalRef\"\u003e10\u003c/span\u003e). This model also predicted higher Ebola cases (3565) from 2019 to 2026, having a 95% confidence interval of 3530 to 3601.\u003c/p\u003e\u003cp\u003e\u003cb\u003eGradient boosting\u003c/b\u003e; there was a greater uncertainty in the prediction of Ebola cases in 2017 and 2018. A high confidence in the prediction of Ebola cases and deaths from 2019 to 2026 and 2017 to 2026 respectively was also registered (Table\u0026nbsp;\u003cspan refid=\"Tab10\" class=\"InternalRef\"\u003e10\u003c/span\u003e). The model predicted higher Ebola cases (3566) and deaths (2365) from 2019 to 2026 having a 95% confidence interval of 3530 to 3603 and 2340 to 2392 respectively.\u003c/p\u003e\u003cp\u003e\u003cb\u003eDeep neural network\u003c/b\u003e; there was a greater uncertainty in the prediction of Ebola cases and deaths from 2017 to 2026. Predictions of Ebola cases and deaths also increased from 2017 to 2026 with higher cases (3556) and deaths (2406) in 2026 having a 95% confidence interval of 1013 to 6102 and 749 to 4117 respectively (Table\u0026nbsp;\u003cspan refid=\"Tab10\" class=\"InternalRef\"\u003e10\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cb\u003eHybrid model\u003c/b\u003e; there was a high confidence in the prediction of Ebola deaths from 2022 to 2026. A greater uncertainty in the prediction of Ebola deaths from 2017 to 2021, 2024 to 2026 was also recorded. Predictions of Ebola deaths by this hybrid model increased from 2017 to 2026 with higher deaths (1590) in 2026 having a 95% confidence interval of 1034 to 2157 (Table\u0026nbsp;\u003cspan refid=\"Tab10\" class=\"InternalRef\"\u003e10\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab10\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 10\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eUncertainty Quantification using the bootstrapping method (CI: Confidence interval)\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"17\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c11\" colnum=\"11\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c12\" colnum=\"12\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c13\" colnum=\"13\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c14\" colnum=\"14\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c15\" colnum=\"15\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c16\" colnum=\"16\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c17\" colnum=\"17\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colspan=\"8\" nameend=\"c9\" namest=\"c2\"\u003e\u003cp\u003eEbola cases\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"8\" nameend=\"c17\" namest=\"c10\"\u003e\u003cp\u003eEbola deaths\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e\u003cp\u003eDecision trees\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003eGradient boosting\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c7\" namest=\"c6\"\u003e\u003cp\u003eDeep Neural Network\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c9\" namest=\"c8\"\u003e\u003cp\u003eHybrid Model\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c11\" namest=\"c10\"\u003e\u003cp\u003eDecision trees\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c13\" namest=\"c12\"\u003e\u003cp\u003eGradient boosting\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c15\" namest=\"c14\"\u003e\u003cp\u003eDeep Neural Network\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c17\" namest=\"c16\"\u003e\u003cp\u003eHybrid Model\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYear\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003eMean\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e95% CI\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003eMean\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e95% CI\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003eMean\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e95% CI\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003eMean\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e\u003cb\u003e95% CI\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e\u003cb\u003eMean\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u003cp\u003e\u003cb\u003e95% CI\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u003cp\u003e\u003cb\u003eMean\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c13\"\u003e\u003cp\u003e\u003cb\u003e95% CI\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c14\"\u003e\u003cp\u003e\u003cb\u003eMean\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c15\"\u003e\u003cp\u003e\u003cb\u003e95% CI\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c16\"\u003e\u003cp\u003e\u003cb\u003eMean\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c17\"\u003e\u003cp\u003e\u003cb\u003e95% CI\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2017\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1737.76\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1570.56, 1913.64\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1737.76\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1570.56, 1913.64\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e2415.08\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e-521.62, 5232.57\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0, 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u003cp\u003e0, 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u003cp\u003e1062.24\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c13\"\u003e\u003cp\u003e959.97, 1173.23\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c14\"\u003e\u003cp\u003e1461.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c15\"\u003e\u003cp\u003e\u0026minus;\u0026thinsp;576.5, 3388.73\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c16\"\u003e\u003cp\u003e841.08\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c17\"\u003e\u003cp\u003e163.62, 1501.42\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2018\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1737.76\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1570.56, 1913.64\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1737.76\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1570.56, 1913.6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e2587.71\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e-417.57, 5566.83\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0, 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u003cp\u003e0, 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u003cp\u003e1062.24\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c13\"\u003e\u003cp\u003e959.97, 1173.23\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c14\"\u003e\u003cp\u003e1617.34\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c15\"\u003e\u003cp\u003e-366.27, 3557.95\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c16\"\u003e\u003cp\u003e893.19\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c17\"\u003e\u003cp\u003e228.72, 1539.18\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2019\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3565.06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3529.7, 3601.37\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e3566.15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e3530.41, 3602.59\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e2784.49\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e-456.02, 6080.08\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0, 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u003cp\u003e0, 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u003cp\u003e2365.33\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c13\"\u003e\u003cp\u003e2339.81, 2391.62\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c14\"\u003e\u003cp\u003e1795.61\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c15\"\u003e\u003cp\u003e\u0026minus;\u0026thinsp;209, 3922.62\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c16\"\u003e\u003cp\u003e1386.98\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c17\"\u003e\u003cp\u003e713.92, 2098.54\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2020\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3565.06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3529.7, 3601.37\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e3566.15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e3530.41, 3602.59\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e3071.06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e163.1, 6110.50\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0, 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u003cp\u003e0, 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u003cp\u003e2365.33\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c13\"\u003e\u003cp\u003e2339.81, 2391.62\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c14\"\u003e\u003cp\u003e1970.35\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c15\"\u003e\u003cp\u003e177.54, 3848.0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c16\"\u003e\u003cp\u003e1445.23\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c17\"\u003e\u003cp\u003e848.37, 2072.16\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2021\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3565.06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3529.7, 3601.37\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e3566.15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e3530.41, 3602.59\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e3298.85\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e714.12, 5778.31\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0, 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u003cp\u003e0, 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u003cp\u003e2365.33\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c13\"\u003e\u003cp\u003e2339.81, 2391.62\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c14\"\u003e\u003cp\u003e2117.95\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c15\"\u003e\u003cp\u003e470.43, 3756.18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c16\"\u003e\u003cp\u003e1494.43\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c17\"\u003e\u003cp\u003e945.84, 2042.02\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2022\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3565.06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3529.7, 3601.37\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e3566.15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e3530.41, 3602.59\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e3494.60\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e2478.33, 4476.06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0, 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u003cp\u003e0, 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u003cp\u003e2365.33\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c13\"\u003e\u003cp\u003e2339.81, 2391.62\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c14\"\u003e\u003cp\u003e2295.80\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c15\"\u003e\u003cp\u003e1675.64, 2975.66\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c16\"\u003e\u003cp\u003e1553.71\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c17\"\u003e\u003cp\u003e1344.58, 1779.75\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2023\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3565.06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3529.7, 3601.37\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e3566.15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e3530.41, 3602.59\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e3542.57\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e2690.79, 4185.72\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0, 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u003cp\u003e0, 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u003cp\u003e2365.33\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c13\"\u003e\u003cp\u003e2339.81, 2391.62\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c14\"\u003e\u003cp\u003e2358.69\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c15\"\u003e\u003cp\u003e1915.2, 2763.27\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c16\"\u003e\u003cp\u003e1574.67\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c17\"\u003e\u003cp\u003e1425.88, 1705.65\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2024\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3565.06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3529.7, 3601.37\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e3566.15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e3530.41, 3602.59\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e3528.57\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e1723.65, 5331.34\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0, 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u003cp\u003e0, 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u003cp\u003e2365.33\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c13\"\u003e\u003cp\u003e2339.81, 2391.62\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c14\"\u003e\u003cp\u003e2384.24\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c15\"\u003e\u003cp\u003e1190.65, 3526.94\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c16\"\u003e\u003cp\u003e1583.19\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c17\"\u003e\u003cp\u003e1188.47, 1965.41\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2025\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3565.06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3529.7, 3601.37\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e3566.15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e3530.41, 3602.59\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e3532.23\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e1347.01, 5708.40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0, 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u003cp\u003e0, 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u003cp\u003e2365.33\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c13\"\u003e\u003cp\u003e2339.81, 2391.62\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c14\"\u003e\u003cp\u003e2395.18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c15\"\u003e\u003cp\u003e970.64, 3826.47\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c16\"\u003e\u003cp\u003e1586.83\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c17\"\u003e\u003cp\u003e1109.47, 2063.22\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2026\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3565.06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3529.7, 3601.37\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e3566.15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e3530.41, 3602.59\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e3556.08\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e1013.24, 6102.24\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0, 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u003cp\u003e0, 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u003cp\u003e2365.33\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c13\"\u003e\u003cp\u003e2339.81, 2391.62\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c14\"\u003e\u003cp\u003e2406.74\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c15\"\u003e\u003cp\u003e748.75, 4116.53\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c16\"\u003e\u003cp\u003e1590.69\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c17\"\u003e\u003cp\u003e1034.21, 2156.95\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eConfidence intervals are represented as (x, y) where x indicates the lower interval and y indicates the upper interval\u003c/p\u003e\u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cdiv id=\"Sec21\" class=\"Section2\"\u003e\u003ch2\u003e4.1 Ebola predictions in Guinea, Liberia, and Sierra Leone\u003c/h2\u003e\u003cp\u003eThe present findings show that in 2014 and from 2015 to 2016, Ebola cases and deaths were exceptionally high in Guinea, Liberia, and Sierra Leone. This may be attributed to factors such as intense deforestation, which are known to increase the human-animal contact, thereby raising the risk of zoonotic disease transmission [\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]. Furthermore, the immense population growth and urbanization, which have enhanced the encroachment on the habitats of wildlife, can be blamed [\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. With the aid of data-driven modeling, Ebola cases and deaths were forecasted to rise from 2016 to 2026 in case of an EVD outbreak in the three countries. However, extremely high Ebola cases and deaths were projected to occur in Sierra Leone than in other countries.\u003c/p\u003e\u003cp\u003eThe possible rise in Ebola cases and deaths from 2016 to 2026 may be attributed to enormous anthropogenic influences such as deforestation. This is possible as deforestation alters the movement patterns and populations of host species [\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e]. Due to deforestation which destroys wildlife habitats, animals such as bats are forced to move into human-inhabited regions formerly unfamiliar to wildlife [\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e], thus fuelling zoonotic outbreaks such as Ebola. This has also been reported in another study where increased spread of Ebola was linked to deforestation [\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e]. Other studies have also reported that pathogens such as viruses are easily transferred from animals to humans, especially if they are around deforested lands [\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eThe urgent need to elevate industrial development in many countries has also driven the destruction of natural habitats for wildlife, thus increasing possible outbreaks of zoonotic diseases [\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e]. The high rate of cross-boundary movements between countries where Ebola has been previously reported may also lead to high Ebola cases and deaths in the future. For example, the recent Ebola outbreaks that have been recorded in the Democratic Republic of Congo, Uganda, and Guinea [\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e] prove it that cross boundary movements are significant contributors of Ebola outbreaks. The rise in Ebola cases and deaths in Guinea, Sierra Leone, and Liberia from 2017 to 2026 may also be linked to changes in global temperatures which are known to influence the emergence of zoonotic diseases such as Ebola [\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e].\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec22\" class=\"Section2\"\u003e\u003ch2\u003e4.2 Characteristics of forecasting models\u003c/h2\u003e\u003cp\u003eOf all the models used, it was the decision trees and deep neural networks that performed better in predicting Ebola cases and deaths. This performance was based on model evaluation and UQ results. In another study, neural networks and decision trees outperformed other forecasting models in predicting COVID-19 cases [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. Other similar studies reported a better performance of neural networks in predicting Parkinson\u0026rsquo;s disease [\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e] and Alzheimer\u0026rsquo;s disease [\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e] than other forecasting models. Another study reported neural network models as performing better than the other models in predicting Monkeypox outbreaks in the USA, Germany, the UK, France, and Canada [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]. The results of the present study where deep neural networks had a better performance after decision trees, are in agreement with a study where neural networks outperformed other models in forecasting [\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e].\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec23\" class=\"Section2\"\u003e\u003ch2\u003e4.3 Limitations of the study, risks of misinterpretation, and integration of intervention scenarios\u003c/h2\u003e\u003cp\u003eMuch as the predictions of Ebola cases and deaths from the four forecasting models are paramount, these predictions can only be relied on in case measures to combat EVD outbreaks are kept constant. This is because measures such as banning of travel to infected communities and early vaccination in infected or suspected communities can help to minimize Ebola cases and deaths. However, in some parts of the world and more so in low-and middle-income countries where resources needed to fully combat zoonotic outbreaks such as Ebola are extremely minimal, the predictions from the current study can be trusted.\u003c/p\u003e\u003cp\u003eThe findings in this study do not necessarily imply that an outbreak of Ebola should be expected in Guinea, Sierra Leone, and Liberia. However, the findings give a clue to the likely number of cases and deaths in case an Ebola outbreak is reported in the three countries. Integrating intervention scenarios such as vaccination campaigns with data-driven infectious disease modelling approaches can altogether help to reduce the predicted number of Ebola cases and deaths in Guinea, Sierra Leone, and Liberia in the face of an outbreak.\u003c/p\u003e\u003c/div\u003e"},{"header":"5. Conclusion and future direction","content":"\u003cp\u003eEbola is one of the deadliest diseases in human history, as it causes significant indirect and direct losses in the communities where it occurs. This underlines the pressing need for timely uncovering, quick reactions, and appropriate public health involvement to combat their spread and effects not only in Guinea, Sierra Leone, and Liberia but also other African countries where outbreaks are likely. Currently, no vaccine for Ebola is fully effective. It is also not easy to accurately predict how severe EVD could be. Due to this, the present research designed a hybrid model using an EVD dataset and compared it with the DT, GB, and DNN models. The time series dataset was obtained from the three mentioned countries. In predicting Ebola cases and deaths, the DT model gave relatively good RMSE (2179.42, 899.17) and R-squared values (0.50, 0.53), and this was followed by the DNN model with RMSE (2195.65, 910.77) and R-squared values (0.49, 0.52). Integrating exogenous covariates such as mobility, demographics, and climate will be pivotal in enhancing forecasting realism. This is because these exogenous covariates are associated with the transmission of zoonotic diseases such as Ebola. Despite the worth of the findings of the present study, future research should focus on the incorporation of data-driven modeling and SIR/SEIR models, as this can help to enhance the present standard epidemiology compartmental models in terms of accuracy.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eEVD\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eEbola virus disease\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eML\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eMachine Learning\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eDT\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eDecision Tree\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eGB\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eGradient Boosting\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eDNN\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eDeep Neural Network\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eANN\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eArtificial Neural Network\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eFI\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eFeature importance\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eCI\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eConfidence Interval\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eSIR\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eSusceptible-Infectious-Recovered\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eSEIR\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eSusceptible-Exposed-Infectious-Recovered\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eUQ\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eUncertainty Quantification\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthical approval\u003c/strong\u003e\u003cp\u003eNot applicable.\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eConsent to participate\u003c/strong\u003e\u003cp\u003eNot applicable.\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eConsent to publish\u003c/strong\u003e\u003cp\u003eThis manuscript has the consent of the author.\u003c/p\u003e\u003c/p\u003e\u003ch2\u003eFunding statement\u003c/h2\u003e\u003cp\u003eThis study did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eThe research concept and design, methodology, drafting of the manuscript, and critical revision of the manuscript for important intellectual content were all done by Thomas James Wanyama.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eI would like to thank Fortunate Amanya for the support and courage provided during the course of this study.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe raw data that was used for predictive modelling is available at [https://www.kaggle.com/datasets/imdevskp/ebola-outbreak-20142016-complete-dataset](https:/www.kaggle.com/datasets/imdevskp/ebola-outbreak-20142016-complete-dataset) .\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003ePigott DM, et al. Mapping the zoonotic niche of Ebola virus disease in Africa. Elife. 2014;3:e04395.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGhareeb OA. Ebola-a fatal emerging zoonotic disease: a review. Ann Rom Soc Cell Biol. 2021;25(6):8748\u0026ndash;54.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJacob ST, et al. Ebola virus disease. Nat Rev Dis Prim. 2020;6(1):13.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTambo E, Ugwu EC, Ngogang JY. Need of surveillance response systems to combat Ebola outbreaks and other emerging infectious diseases in African countries. Infect Dis poverty. 2014;3:1\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLokko CNM, Lokko FO. EPIDEMIOLOGY OF HUMAN EBOLA VIRUS DISEASE OUTBREAK IN SUB-SAHARAN AFRICA, 2025.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLokko CNAAM, Lokko FO, HUMAN EBOLA VIRUS DISEASE OUTBREAK IN SUB SAHARAN AFRICA. : IMPLICATION FOR BORDER TOWNS ACROSS THE GLOBE, 2016.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAlexander KA, et al. What factors might have led to the emergence of Ebola in West Africa? PLoS Negl Trop Dis. 2015;9(6):e0003652.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eColtart CEM, Lindsey B, Ghinai I, Johnson AM, Heymann DL. The Ebola outbreak, 2013\u0026ndash;2016: old lessons for new epidemics. Philos Trans R Soc B Biol Sci. 2017;372(1721):20160297.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWhite RJ, Razgour O. Emerging zoonotic diseases originating in mammals: a systematic review of effects of anthropogenic land-use change. Mamm Rev. 2020;50(4):336\u0026ndash;52.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWHO. Regional Office for Africa. In Africa, 63% jump in diseases spread from animals to people seen in last decade, 2024.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHashem S, et al. Machine learning prediction models for diagnosing hepatocellular carcinoma with HCV-related chronic liver disease. Comput Methods Programs Biomed. 2020;196:105551.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChowdhury MEH, et al. An early warning tool for predicting mortality risk of COVID-19 patients using machine learning. Cognit Comput. 2024;16(4):1778\u0026ndash;93.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMurthy RK. Early detection and prediction of zoonotic disease events using event-based surveillance and machine learning. Washington State University; 2023.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChen S, Ding Y. The feasibility of using machine learning to predict COVID-19 cases. Int J Med Inf. 2025;196:105786. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ijmedinf.2025.105786\u003c/span\u003e\u003cspan address=\"10.1016/j.ijmedinf.2025.105786\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang P, Zheng X, Ai G, Liu D, Zhu B. Time series prediction for the epidemic trends of COVID-19 using the improved LSTM deep learning method: Case studies in Russia, Peru and Iran. Chaos Solitons Fractals. 2020;140:110214.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eXia Z, Qin L, Ning Z, Zhang X. Deep learning time series prediction models in surveillance data of hepatitis incidence in China. PLoS ONE. 2022;17(4):e0265660.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMcNamara LA. Ebola Surveillance\u0026mdash;Guinea, Liberia, and Sierra Leone. MMWR Suppl, 65, 2016.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eShaman J, Yang W, Kandula S. Inference and forecast of the current West African Ebola outbreak in Guinea, Sierra Leone and Liberia. PLoS Curr. 2014;6:ecurrents\u0026ndash;outbreaks.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMeltzer MI, et al. Estimating the future number of cases in the Ebola epidemic\u0026mdash;Liberia and Sierra Leone, 2014\u0026ndash;2015. MMWR Surveill Summ. 2014;63:1\u0026ndash;14.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLiu A. Diagnosis and Prognosis Prediction for Ebola Virus Disease Using Machine Learning Methods, 2021.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMannan A, Shoket N, Rahman JU, Uwitije R. Dynamic analysis of ebola virus disease with non-linear incidence rate using morlet wavelet neural networks and hybrid optimization techniques. Model Earth Syst Environ. 2025;11(2):79.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDou Y, Meng W. Predictive Analysis of The Number of Ebola Virus Infections Based on Grey Prediction Model and Neural Network, in \u003cem\u003e2022 IEEE 10th Joint International Information Technology and Artificial Intelligence Conference (ITAIC)\u003c/em\u003e, IEEE, 2022, pp. 190\u0026ndash;195.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSreenivasan MA, Bhat HR, Rajagopalan PK. The epizootics of Kyasanur Forest disease in wild monkeys during 1964 to 1973. Trans R Soc Trop Med Hyg. 1986;80(5):810\u0026ndash;4.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKumar A, Saxena N, Jung S, Choi BJ. Improving detection of false data injection attacks using machine learning with feature selection and oversampling. Energies. 2021;15(1):212.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWiens M, Verone-Boyle A, Henscheid N, Podichetty JT, Burton J. A tutorial and use case example of the eXtreme gradient boosting (XGBoost) artificial intelligence algorithm for drug development applications. Clin Transl Sci. 2025;18(3):e70172.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eUpadhyay D, Manero J, Zaman M, Sampalli S. Gradient boosting feature selection with machine learning classifiers for intrusion detection on power grids. IEEE Trans Netw Serv Manag. 2020;18(1):1104\u0026ndash;16.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYe L, Jabbar S, Abdul Zahra M, Tan ML. Bayesian Regularized Neural Network Model Development for Predicting Daily Rainfall from Sea Level Pressure Data: Investigation on Solving Complex Hydrology Problem, \u003cem\u003eComplexity\u003c/em\u003e, vol. 2021, pp. 1\u0026ndash;14, Apr. 2021. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1155/2021/6631564\u003c/span\u003e\u003cspan address=\"10.1155/2021/6631564\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang C, Xu J, Tan S, Yin L. Secure decision tree classification with decentralized authorization and access control. Comput Stand Interfaces. 2024;89:103818.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMienye ID, Jere N. A survey of decision trees: Concepts, algorithms, and applications. IEEE access. 2024;12:86716\u0026ndash;27.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRivera-Lopez R, Canul-Reich J, Mezura-Montes E, Cruz-Ch\u0026aacute;vez MA. Induction of decision trees as classification models through metaheuristics. Swarm Evol Comput. 2022;69:101006.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eItzkin M, Palmsten ML, Buckley ML, Birchler JJ, Torres-Garcia LM. Developing a decision tree model to forecast runup and assess uncertainty in empirical formulations. Coast Eng. 2025;195:104641. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.coastaleng.2024.104641\u003c/span\u003e\u003cspan address=\"10.1016/j.coastaleng.2024.104641\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJi XC, et al. Recent advances in machine learning for defects detection and prediction in laser cladding process. Next Mater. 2025;7:100404. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.nxmate.2024.100404\u003c/span\u003e\u003cspan address=\"10.1016/j.nxmate.2024.100404\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAtiku S, Obagbuwa I. Machine Learning Classification Techniques for Detecting the Impact of Human Resources Outcomes on Commercial Banks Performance, \u003cem\u003eAppl. Comput. Intell. Soft Comput.\u003c/em\u003e, vol. 2021, pp. 1\u0026ndash;16, Sep. 2021. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1155/2021/7747907\u003c/span\u003e\u003cspan address=\"10.1155/2021/7747907\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMingard C, Rees H, Valle-P\u0026eacute;rez G, Louis AA. Deep neural networks have an inbuilt Occam\u0026rsquo;s razor. Nat Commun. 2025;16(1):220.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRajesh G, Raajini XM, Dang H. Industry 4.0 interoperability, analytics, security, and case studies. CRC; 2021.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePeterson DA, Sejnowski TJ. A dynamic circuit hypothesis for the pathogenesis of blepharospasm. Front Comput Neurosci. 2017;11:11.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMehlig B. Machine learning with neural networks: an introduction for scientists and engineers. Cambridge University Press; 2021.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCao W, Wang X, Ming Z, Gao J. A review on neural networks with random weights. Neurocomputing. 2018;275:278\u0026ndash;87.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eManohar B, Das R. Artificial neural networks for the prediction of monkeypox outbreak. Trop Med Infect Dis. 2022;7(12):424.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eArrieta AB, et al. Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI. Inf fusion. 2020;58:82\u0026ndash;115.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRengasamy D, et al. Feature importance in machine learning models: A fuzzy information fusion approach. Neurocomputing. 2022;511:163\u0026ndash;74.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eArdabili SF, Mahmoudi A, Gundoshmian TM. Modeling and simulation controlling system of HVAC using fuzzy and predictive (radial basis function, RBF) controllers. J Build Eng. 2016;6:301\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJalaian B, Lee M, Russell S. Uncertain context: Uncertainty quantification in machine learning. AI Mag. 2019;40(4):40\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFigueroa D, Duprat X. Remedying anthropogenic zoonoses. Anim Sentience. 2020;5:29.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTian H et al. Urbanization prolongs hantavirus epidemics in cities, \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e, vol. 115, no. 18, pp. 4707\u0026ndash;4712, 2018.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRochlin I, Egizi A, Lindstr\u0026ouml;m A. The original scientific description of the lone star tick (Amblyomma americanum, Acari: Ixodidae) and implications for the species\u0026rsquo; past and future geographic distributions. J Med Entomol. 2022;59(2):412\u0026ndash;20.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRulli MC, Santini M, Hayman DTS, D\u0026rsquo;Odorico P. The nexus between forest fragmentation in Africa and Ebola virus disease outbreaks. Sci Rep. 2017;7(1):41613.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eOlivero J, et al. Recent loss of closed forests is associated with Ebola virus disease outbreaks. Sci Rep. 2017;7(1):14291.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMorand S, Lajaunie C. Outbreaks of vector-borne and zoonotic diseases are associated with changes in forest cover and oil palm expansion at global scale. Front Vet Sci. 2021;8:661063.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVinson JE, et al. Land reversion and zoonotic spillover risk. R Soc open Sci. 2022;9(6):220582.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMoyo E, Mhango M, Moyo P, Dzinamarira T, Chitungo I, Murewanhema G. Emerging infectious disease outbreaks in Sub-Saharan Africa: Learning from the past and present to be better prepared for future outbreaks. Front public Heal. 2023;11:1049986.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSachan N, Singh VP. Effect of climatic changes on the prevalence of zoonotic diseases. Vet World. 2010;3(11):519.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGuimar\u0026atilde;es P, Keller A, Fehlmann T, Lammert F, Casper M. Deep-learning based detection of gastric precancerous conditions. Gut. 2020;69(1):4\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTabarestani S et al. Longitudinal Prediction Modeling of Alzheimer Disease using Recurrent Neural Networks, in., 2019 \u003cem\u003eIEEE EMBS International Conference on Biomedical \u0026amp; Health Informatics (BHI)\u003c/em\u003e, 2019, pp. 1\u0026ndash;4. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1109/BHI.2019.8834556\u003c/span\u003e\u003cspan address=\"10.1109/BHI.2019.8834556\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMasini RP, Medeiros MC, Mendes EF. Machine learning advances for time series forecasting. J Econ Surv. 2023;37(1):76\u0026ndash;111.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Decision Tree, Deep Neural Network, Ebola virus Disease, Gradient Boosting, Data-driven Modeling, Uncertainty quantification","lastPublishedDoi":"10.21203/rs.3.rs-6969004/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6969004/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eEbola is among the deadliest human diseases and some measures to combat it have been developed, but they are still not able to fully minimize its far-reaching effects. However, the integration of data-driven modeling into existing Ebola prevention and control measures offers hope toward combating this viral disease. Decision trees (DT), gradient boosting (GB), deep neural networks (DNN), and a hybrid model were used to predict Ebola cases and deaths in Guinea, Liberia, and Sierra Leone from 2017 to 2026. These models were evaluated using mean absolute scaled error (MASE), root mean squared error (RMSE), and coefficient of determination (R\u0026sup2;). The uncertainties in the prediction of the models were also quantified using the bootstrapping method. Of all the models, the DT model had higher feature importance scores for cases in two countries. The DT model also had a higher accuracy and better performance by RMSE and R\u003csup\u003e2\u003c/sup\u003e in predicting Ebola cases and deaths than other models. Predictions of Ebola cases and deaths by the DNN model increased from 2017 to 2026 with higher cases and deaths in 2026 while predictions of Ebola deaths by the hybrid model increased from 2017 to 2026 with higher deaths in 2026. The projected Ebola cases and deaths were higher in Sierra Leone than in other countries. These findings portray the likely number of cases and deaths in case an Ebola outbreak in the three mentioned countries. Furthermore, they show their significance in predicting Ebola virus disease and also have the possibility to help decision-makers in designing effective decisions for the early detection of Ebola incidents. The results of this study show that the DT and DNN models perform better than the other models on the collected Ebola virus disease dataset in the three countries. Therefore, the integration of data-driven infectious disease modeling approaches such as DT and DNN with intervention scenarios such as vaccination can altogether help to reduce the predicted number of Ebola cases and deaths in Guinea, Sierra Leone, and Liberia in the face of an outbreak.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e","manuscriptTitle":"Data-driven Modeling as a Tool for Prediction of Future Outbreaks of Ebola Virus Disease in West Africa","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-01 08:04:38","doi":"10.21203/rs.3.rs-6969004/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"05ce086d-4965-44e6-b6e9-44a7fdbd3615","owner":[],"postedDate":"October 1st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-10-01T08:04:39+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-01 08:04:38","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6969004","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6969004","identity":"rs-6969004","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}