Machine Learning Approaches for Temporal and Spatio-Temporal Covid-19 Forecasting: A Brief Review and a Contribution

In December 2019 an outbreak of a new disease happened, in Wuhan city, China, in which the symptoms were very similar to pneumonia. The disease was attributed to SARS-CoV-2 as the infectious agent and it was called the new coronavirus or Covid-19. In March 2020, the World Health Organization declared a worldwide pandemic of the new coronavirus. We have already counted more than 110 million cases and almost 2.5 million deaths worldwide. In order to assist in decision-making to contain the disease, several scientists around the world have engaged in various efforts, and they have proposed a lot of systems and solutions for tracking, monitoring, and predicting confirmed cases and deaths from Covid-19. Mathematical models help to analyze and understand the evolution of the disease, but understanding the disease was not enough, it was necessary to understand the problem in a quantitative way to lead the decision-making during the pandemic. Several initiatives have made use of Artificial Intelligence, and models were designed using machine learning algorithms with features for temporal and spatio-temporal investigation and prediction of cases of Covid-19. Among the algorithms used are Support Vector Machine (SVM), Random Forest, Multilayer Perceptron (MLP), Graph Neural Networks (GNNs), Ecological Niche Models (ENMs), Long-Short Term Memory Networks (LSTM), linear regression, and others. And these had good results, and to analyze them, the Root Mean Squared Error (RMSE), Log Root Mean Squared Error (RMSLE), correlation coefficient, and others were used as metrics. Covid-19 presents a huge problem to public health worldwide, so it is of utmost importance to investigate it, and with these two approaches it is possible to track not only how the disease evolves but also to know which areas are at risk. And these solutions can help in supporting decision-making by health managers to make the best decisions for the disease that is in the outbreak. This chapter aims to present a literature review and a brief contribution to the use of machine learning methods for temporal and spatio-temporal prediction of Covid-19, using Brazil and its federative units as a case study. From canonical methods to deep networks and hybrid committee-based, approaches will be investigated. © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2022.

Covid-19 Forecasting Using CNN Approach With A Halbinomial Distribution And A Linear Decreasing Inertia Weight-Based Cat Swarm Optimization

In recent times, the COVID-19 epidemic has spread to over 170 nations. Authorities all around the world are feeling the strain of COVID-19 since the total of infected people is rising as well as they does not familiar to handle the problem. The majority of current research effort is thus being directed on the analysis of COVID-19 data within the framework of the machines learning method. Researchers looked the COVID 19 data to make predictions about who would be treated, who would die, and who would get infected in the future. This might prompt governments worldwide to develop strategies for protecting the health of the public. Previous systems rely on Long Short -Term Memory (LSTM) networks for predicting new instances of COVID-19. The LSTM network findings suggest that the pandemic might be over by June of 2020. However, LSTM may have an over-fitting issue, and it may fall short of expectations in terms of true positive. For this issue in COVID-19 forecasting, we suggest using two methods such as Cat Swarm Optimization (CSO) for reducing the inertia weight linearly and then artificial intelligence based binomial distribution is used. In this proposed study, we take the COVID-19 predicting database as an contribution and normalise it using the min-max approach. The accuracy of classification is improved with the use of the first method to choose the optimal features. In this method, inertia weight is added to the CSO optimization algorithm convergence. Death and confirmed cases are predicted for a certain time period throughout India using Convolutional Neural Network with Partial Binomial Distribution based on carefully chosen characteristics. The experimental findings validate that the suggested scheme performs better than the baseline system in terms of f-measure, recall, precision, and accuracy.

A Multivariate Deep Learning Model with Coupled Human Intervention Factors for COVID-19 Forecasting

Artificial intelligence (AI) technology plays a crucial role in infectious disease outbreak prediction and control. Many human interventions can influence the spread of epidemics, including government responses, quarantine, and economic support. However, most previous AI-based models have failed to consider human interventions when predicting the trend of infectious diseases. This study selected four human intervention factors that may affect COVID-19 transmission, examined their relationship to epidemic cases, and developed a multivariate long short-term memory network model (M-LSTM) incorporating human intervention factors. Firstly, we analyzed the correlations and lagged effects between four human factors and epidemic cases in three representative countries, and found that these four factors typically delayed the epidemic case data by approximately 15 days. On this basis, a multivariate epidemic prediction model (M-LSTM) was developed. The model prediction results show that coupling human intervention factors generally improves model performance, but adding certain intervention factors also results in lower performance. Overall, a multivariate deep learning model with coupled variable correlation and lag outperformed other comparative models, and thus validated its effectiveness in predicting infectious diseases.

Forecasting COVID-19 Cases Using n-SARS-CoV-2 Variants

This work proposes a novel Deep Learning-based model to forecast the total number of confirmed COVID-19 cases in four of the worst-hit states of India. Along with statewide restrictions and public holidays, a novel parameter is introduced for training the proposed model, which considers the Alpha, Beta, Delta, and Omicron variants and the degree of their prevalence in each of the four states. Recurrent Neural Network-based Long-Short Term Memory is applied to the custom dataset, with the lowest Mean Absolute Percentage Error being 0.77% for the state of Maharashtra. SHapley Additive exPlanations values are used to examine the significance of the various parameters. The proposed model can be applied to other countries and can include newer variants of the novel coronavirus discovered in the future. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Forecasting COVID-19 and Analyzing the Effect of Government Interventions

We developed DELPHI, a novel epidemiological model for predicting detected cases and deaths in the prevaccination era of the COVID-19 pandemic. The model allows for underdetection of infections and effects of government interventions. We have applied DELPHI across more than 200 geographical areas since early April 2020 and recorded 6% and 11% two-week, out-of-sample median mean absolute percentage error on predicting cases and deaths, respectively. DELPHI compares favorably with other top COVID-19 epidemiological models and predicted in 2020 the large-scale epidemics in many areas, including the United States, United Kingdom, and Russia, months in advance. We further illustrate two downstream applications of DELPHI, enabled by the model's flexible parametric formulation of the effect of government interventions. First, we quantify the impact of government interventions on the pandemic's spread. We predict, that in the absence of any interventions, more than 14 million individuals would have perished by May 17, 2020, whereas 280,000 deaths could have been avoided if interventions around the world had started one week earlier. Furthermore, we find that mass gathering restrictions and school closings were associated with the largest average reductions in infection rates at 29.9±6.9% and 17.3±6.7% , respectively. The most stringent policy, stay at home, was associated with an average reduction in infection rate by 74.4±3.7% from baseline across countries that implemented it. In the second application, we demonstrate how DELPHI can predict future COVID-19 incidence under alternative governmental policies and discuss how Janssen Pharmaceuticals used such analyses to select the locations of its Phase III trial for its leading single-dose vaccine candidate Ad26.Cov2.S.

Will gold prices persist post pandemic period? An econometric evidence

Recurrent stock market fall and rise sequel by COVID-19, rising global inflation, increase in Fed interest rates, the unprecedented meltdown of technology stocks, fear of trade wars, tightening of governments' fiscal policies call for a new trend in international investing. It is time for the investors to rethink, rebalance and reset their investment strategies to position and protect their portfolios during and post-pandemic period. This paper attempts to forecast the gold prices for the post-pandemic era and explores whether gold will serve as a decisive hedge during this transition period. The techniques of ARCH, GARCH, E-GARCH, A-PARCH, and GARCH-M is employed in forecasting the conditional volatility of gold spot price from Multi Commodity Exchange (MCX) of India. A total of 3631 observations were collected from the daily spot prices of gold from January 2009 to December 2022. The findings show that the gold prices in India are highly persistent similar to other emerging markets and that gold will remain a safe haven for investors and institutional investors in the post-pandemic period. This paper is the first of its kind to forecast gold prices for the post-pandemic period. The forecast price of 10-gram gold is expected to trade for 65,948 ₹ in the Indian MCX by 2026 if the gold prices behold its previous momentum. This forecast will help the investors to plan their portfolio diversification for the post-pandemic period. © 2023, MDPI. All rights reserved.

Nonparametric Homogeneity Test for Small Samples and Its Applications to COVID-19 Epidemic Curves Forecasting

The COVID-19 pandemic is accompanied by intensive attempts to build mathematical models to predict it. For this, various models are used, both traditional differential equations and machine learning models. Classical epidemiological compartment models contain parameters that are difficult to measure. Their results are used to model various scenarios, but it is difficult to obtain a reliable forecast with their help. Machine learning models, on the other hand, do not use prior assumptions, and their inferences are based only on training samples. This usually results in more reliable forecasts. In both the first and second cases, it is necessary not only to estimate the forecast error, but to compare the prediction accuracy of different models by checking the error homogeneity also. An additional factor complicating the problem is the small size of available samples in some cases. This forces one to resort to resampling methods. The article describes the Klyushin-Petunin test for testing the homogeneity of samples with ties in a multi-sample design and compares it with the traditional Anderson-Darling, Kruskal-Wallis and Friedman tests using the example of three methods for predicting the COVID-19 epidemic in the basis of epidemic data in Germany, Japan, South Korea and Ukraine. © 2022 IEEE.

COVID-19 Forecasting Based on Local Mean Decomposition and Temporal Convolutional Network

Since the outbreak of coronavirus disease 2019 (COVID-19) has resulted in a dramatic loss of human life and economic disruption worldwide from early 2020, numerous studies focusing on COVID-19 forecasting were presented to yield accurate predicting results. However, most existing methods could not provide satisfying forecasting performance due to tons of assumptions, poor capability to learn appropriate parameters, etc. Therefore, in this paper, we combine a traditional time series decomposition: local mean decomposition (LMD) with temporal convolutional network (TCN) as a general framework to overcome these shortcomings. Based on the particular architecture, it can solve weekly new confirmed cases forecasting problem perfectly. Extensive experiments show that the proposed model significantly outperforms lots of state-of-the-art forecasting methods, and achieves desirable performance in terms of root mean squared log error (RMSLE), mean absolute percentage error (MAPE), Pearson correlation (PCORR), and coefficient of determination (R-2). To be specific, it could reach 0.9739, 0.8908, and 0.7461 on R-2 when horizon is 1, 2, and 3 respectively, which proves the effectiveness and robustness of our LMD-TCN model.

Time series analysis of the spread of COVID-19 infection in Bulgaria

The number of confirmed, deaths and recovered cases of Covid-19 in.the period 08.03.2020-29.04.2021 in.Bulgaria are considered. Various smoothing techniques have been used to highlight the trend in data change. Confidence intervals are obtained for daily and weekly deaths after 14 days based on confirmed cases to date. The Box-Cox transformed confirmed cases were modeled by regression with respect to three dummy variables related to working days and holidays, with errors following ARIMA(1,1,2)(2,0,0)7 model. For Box-Cox transformed deaths, three other dummy variables and lag 16 of transformed and smoothed confirmed cases were used. The regression errors follow ARIMA(1,1,2) model in which the MA(1) coefficient is set to 0. Fourteen daily forecasts are obtained which agree well with the respective test data sets. © 2022 Author(s).

Spatial-Temporal Synchronous Graph Transformer network (STSGT) for COVID-19 forecasting.

COVID-19 has become a matter of serious concern over the last few years. It has adversely affected numerous people around the globe and has led to the loss of billions of dollars of business capital. In this paper, we propose a novel Spatial-Temporal Synchronous Graph Transformer network (STSGT) to capture the complex spatial and temporal dependency of the COVID-19 time series data and forecast the future status of an evolving pandemic. The layers of STSGT combine the graph convolution network (GCN) with the self-attention mechanism of transformers on a synchronous spatial-temporal graph to capture the dynamically changing pattern of the COVID time series. The spatial-temporal synchronous graph simultaneously captures the spatial and temporal dependencies between the vertices of the graph at a given and subsequent time-steps, which helps capture the heterogeneity in the time series and improve the forecasting accuracy. Our extensive experiments on two publicly available real-world COVID-19 time series datasets demonstrate that STSGT significantly outperforms state-of-the-art algorithms that were designed for spatial-temporal forecasting tasks. Specifically, on average over a 12-day horizon, we observe a potential improvement of 12.19% and 3.42% in Mean Absolute Error (MAE) over the next best algorithm while forecasting the daily infected and death cases respectively for the 50 states of US and Washington, D.C. Additionally, STSGT also outperformed others when forecasting the daily infected cases at the state level, e.g., for all the counties in the State of Michigan. The code and models are publicly available at https://github.com/soumbane/STSGT.

Machine Learning Approach for Forecast Analysis of Novel COVID-19 Scenarios in India

The novel coronavirus (nCOV) is a new strain that needs to be hindered from spreading by taking effective preventive measures as swiftly as possible. Timely forecasting of COVID-19 cases can ultimately support in making significant decisions and planning for implementing preventive measures. In this study, three common machine learning (ML) approaches via linear regression (LR), sequential minimal optimization (SMO) regression, and M5P techniques have been discussed and implemented for forecasting novel coronavirus disease-2019 (COVID-19) pandemic scenarios. To demonstrate the forecast accuracy of the aforementioned ML approaches, a preliminary sample-study has been conducted on the first wave of the COVID-19 pandemic scenario for three different countries including the United States of America (USA), Italy, and Australia. Furthermore, the contributions of this study are extended by conducting an in-depth forecast study on COVID-19 pandemic scenarios for the first, second, and third waves in India. An accurate forecasting model has been proposed, which has been constructed on the basis of the results of the aforementioned forecasting models of COVID-19 pandemic scenarios. The findings of the research highlight that LR is a potential approach that outperforms all other forecasting models tested herein in the present COVID-19 pandemic scenario. Finally, the LR approach has been used to forecast the likely onset of the fourth wave of COVID-19 in India.

Multiwave COVID-19 Prediction from Social Awareness Using Web Search and Mobility Data

Recurring outbreaks of COVID-19 have posed enduring effects on global society, which calls for a predictor of pandemic waves using various data with early availability. Existing prediction models that forecast the first outbreak wave using mobility data may not be applicable to the multiwave prediction, because the evidence in the USA and Japan has shown that mobility patterns across different waves exhibit varying relationships with fluctuations in infection cases. Therefore, to predict the multiwave pandemic, we propose a Social Awareness-Based Graph Neural Network (SAB-GNN) that considers the decay of symptom-related web search frequency to capture the changes in public awareness across multiple waves. Our model combines GNN and LSTM to model the complex relationships among urban districts, inter-district mobility patterns, web search history, and future COVID-19 infections. We train our model to predict future pandemic outbreaks in the Tokyo area using its mobility and web search data from April 2020 to May 2021 across four pandemic waves collected by Yahoo Japan Corporation under strict privacy protection rules. Results demonstrate our model outperforms state-of-the-art baselines such as ST-GNN, MPNN, and GraphLSTM. Though our model is not computationally expensive (only 3 layers and 10 hidden neurons), the proposed model enables public agencies to anticipate and prepare for future pandemic outbreaks. © 2022 Owner/Author.

An innovative ensemble model based on multiple neural networks and a novel heuristic optimization algorithm for COVID-19 forecasting.

During the global fight against the novel coronavirus pneumonia (COVID-19) epidemic, accurate outbreak trend forecasting has become vital for outbreak prevention and control. Effective COVID-19 outbreak trend prediction remains a complex and challenging issue owing to the significant fluctuations in the COVID-19 data series. Most previous studies have limitations only using individual forecasting methods for outbreak modeling, ignoring the combination of the advantages of different prediction methods, which may lead to insufficient results. Therefore, this paper develops a novel ensemble paradigm based on multiple neural networks and a novel heuristic optimization algorithm. First, a new hybrid sine cosine algorithm-whale optimization algorithm (SCWOA) is exercised on 15 benchmark tests. Second, four neural networks are used as predictors for the COVID-19 outbreak forecasting. Each predictor is given a weight, and the proposed SCWOA is used to optimize the best matching weights of the ensemble model. The daily COVID-19 series collected from three of the most-affected countries were taken as the test cases. The experimental results demonstrate that different neural network models have different performances in various complex epidemic prediction scenarios. The SCWOA-based ensemble model can outperform all comparable models with its high accuracy and robustness.

Covid-19 Forecasting with Deep Learning-based Half-binomial Distribution Cat Swarm Optimization

About 170 nations have been affected by the COvid VIrus Disease-19 (COVID-19) epidemic. On governing bodies across the globe, a lot of stress is created by COVID-19 as there is a continuous rise in patient count testing positive, and they feel challenging to tackle this situation. Most researchers concentrate on COVID-19 data analysis using the machine learning paradigm in these situations. In the previous works, Long Short-Term Memory (LSTM) was used to predict future COVID-19 cases. According to LSTM network data, the outbreak is expected to finish by June 2020. However, there is a chance of an over-fitting problem in LSTM and true positive;it may not produce the required results. The COVID-19 dataset has lower accuracy and a higher error rate in the existing system. The proposed method has been introduced to overcome the above-mentioned issues. For COVID-19 prediction, a Linear Decreasing Inertia Weight-based Cat Swarm Optimization with Half Binomial Distribution based Convolutional Neural Network (LDIWCSO-HBDCNN) approach is presented. In this suggested research study, the COVID-19 predicting dataset is employed as an input, and the min-max normalization approach is employed to normalize it. Optimum features are selected using Linear Decreasing Inertia Weight-based Cat Swarm Optimization (LDIWCSO) algorithm, enhancing the accuracy of classification. The Cat Swarm Optimization (CSO) algorithm???s convergence is enhanced using inertia weight in the LDIWCSO algorithm. It is used to select the essential features using the best fitness function values. For a specified time across India, death and confirmed cases are predicted using the Half Binomial Distribution based Convolutional Neural Network (HBDCNN) technique based on selected features. As demonstrated by empirical observations, the proposed system produces significant performance in terms of f-measure, recall, precision, and accuracy.

Deep Spatiotemporal Model for COVID-19 Forecasting.

COVID-19 has caused millions of infections and deaths over the last 2 years. Machine learning models have been proposed as an alternative to conventional epidemiologic models in an effort to optimize short- and medium-term forecasts that will help health authorities to optimize the use of policies and resources to tackle the spread of the SARS-CoV-2 virus. Although previous machine learning models based on time pattern analysis for COVID-19 sensed data have shown promising results, the spread of the virus has both spatial and temporal components. This manuscript proposes a new deep learning model that combines a time pattern extraction based on the use of a Long-Short Term Memory (LSTM) Recurrent Neural Network (RNN) over a preceding spatial analysis based on a Convolutional Neural Network (CNN) applied to a sequence of COVID-19 incidence images. The model has been validated with data from the 286 health primary care centers in the Comunidad de Madrid (Madrid region, Spain). The results show improved scores in terms of both root mean square error (RMSE) and explained variance (EV) when compared with previous models that have mainly focused on the temporal patterns and dependencies.

Stochastic forecasting of COVID-19 daily new cases across countries with a novel hybrid time series model.

An unprecedented outbreak of the novel coronavirus (COVID-19) in the form of peculiar pneumonia has spread globally since its first case in Wuhan province, China, in December 2019. Soon after, the infected cases and mortality increased rapidly. The future of the pandemic's progress was uncertain, and thus, predicting it became crucial for public health researchers. These predictions help the effective allocation of health-care resources, stockpiling, and help in strategic planning for clinicians, government authorities, and public health policymakers after understanding the extent of the effect. The main objective of this paper is to develop a hybrid forecasting model that can generate real-time out-of-sample forecasts of COVID-19 outbreaks for five profoundly affected countries, namely the USA, Brazil, India, the UK, and Canada. A novel hybrid approach based on the Theta method and autoregressive neural network (ARNN) model, named Theta-ARNN (TARNN) model, is developed. Daily new cases of COVID-19 are nonlinear, non-stationary, and volatile; thus, a single specific model cannot be ideal for future prediction of the pandemic. However, the newly introduced hybrid forecasting model with an acceptable prediction error rate can help healthcare and government for effective planning and resource allocation. The proposed method outperforms traditional univariate and hybrid forecasting models for the test datasets on an average.

Study on the Dynamics of COVID-19 Cases in Achieving Herd Immunity in Indonesia

Group immunity or herd immunity is a crucial condition that determines whether or not the COVID-19 outbreak is controlled or not. Government policies, both in terms of social control and vaccination, are one of the important factors in achieving group immunity. In this paper, an analysis of the dynamics of COVID-19 cases in Indonesia is carried out in correlation with government policies and also the rate of vaccination. We found that vaccination is the most important key in achieving group immunity and this will lead to Indonesian mobility behavior towards COVID-19 from time to time. Government Policies also play a significant effort toward vaccinations starting from the beginning (PSBB) to Emergency PPKM. This study is not considered a new variant that is resistant against vaccines, it may take more time in achieving group immunity if the new variants exist. This analysis leads to a deduction of the time required for Indonesia to achieve herd immunity. This study also estimates the time series of cases and vaccinations using the N-Beats model to strengthen the deductions made from past dynamics. Based on this study, it is estimated that in February 2022 a mask removal policy will be issued and in October 2021 COVID-19 positive cases will be declined. © 2021 IEEE.

COVID-19 Epidemic Process Simulation Using ARIMA Model

The substantial ascendant trend within the number of daily infected new cases with coronavirus around the world is a warning, and several other researchers are utilizing various mathematical and machine learning-based prediction models to forecast the long-term trend of the COVID-19 pandemic. During this research, the Autoregressive Integrated Moving Average or ARIMA model was implemented to forecast the COVID-19 expected daily number of cases in Ukraine. We implemented Autoregressive Integrated Moving Average for this research. The forecasting results showed that the trend in Ukraine will continue ascending and should reach up to more than 1.8 million total cases if stringent precautionary and control measures don’t get implemented to prohibit the spread of COVID-19. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Interpretable Temporal Attention Network for COVID-19 forecasting.

The worldwide outbreak of coronavirus disease 2019 (COVID-19) has triggered an unprecedented global health and economic crisis. Early and accurate forecasts of COVID-19 and evaluation of government interventions are crucial for governments to take appropriate interventions to contain the spread of COVID-19. In this work, we propose the Interpretable Temporal Attention Network (ITANet) for COVID-19 forecasting and inferring the importance of government interventions. The proposed model is with an encoder-decoder architecture and employs long short-term memory (LSTM) for temporal feature extraction and multi-head attention for long-term dependency caption. The model simultaneously takes historical information, a priori known future information, and pseudo future information into consideration, where the pseudo future information is learned with the covariate forecasting network (CFN) and multi-task learning (MTL). In addition, we also propose the degraded teacher forcing (DTF) method to train the model efficiently. Compared with other models, the ITANet is more effective in the forecasting of COVID-19 new confirmed cases. The importance of government interventions against COVID-19 is further inferred by the Temporal Covariate Interpreter (TCI) of the model.

Multi-modal Information Fusion-powered Regional Covid-19 Epidemic Forecasting

With the current raging spread of the COVID19, early forecasting of the future epidemic trend is of great significance to public health security. The COVID-19 is virulent and spreads widely. An outbreak in one region often triggers the spread of others, and regions with relatively close association would show a strong correlation in the spread of the epidemic. In the real world, many factors affect the spread of the outbreak between regions. These factors exist in the form of multimodal data, such as the time-series data of the epidemic, the geographic relationship, and the strength of social contacts between regions. However, most of the current work only uses historical epidemic data or single-modal geographic location data to forecast the spread of the epidemic, ignoring the correlation and complementarity in multi-modal data and its impact on the disease spread between regions. In this paper, we propose a Multimodal InformatioN fusion COVID-19 Epidemic forecasting model (MINE). It fuses inter-regional and intra-regional multi-modal information to capture the temporal and spatial relevance of the COVID-19 spread in different regions. Extensive experimental results show that the proposed method achieves the best results compared to state-of-art methods on benchmark datasets. © 2021 IEEE.