Fig 7 - uploaded by Junaid Farooq
Content may be subject to copyright.
30-day Forecast for Maharashtra: Red line shows the start of prediction window. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Source publication
In this paper, deep learning is employed to propose an Artificial Neural Network (ANN) based online incremental learning technique for developing an adaptive and non-intrusive analytical model of Covid-19 pandemic to analyze the temporal dynamics of the disease spread. The model is able to intelligently adapt to new ground realities in real-time el...
Contexts in source publication
Context 1
... in Mumbai, is of particular concern as it is considered to be one of Asia's largest slums with an area of just over 2.1 square kilometres and a population of about 1,000,000. Therefore, the rules of social distancing are difficult to observe there. Further, Mumbai has shortage of ICU beds and dedicated COVID-19 hospitals as well. As shown in Fig. 7, Maharashtra is expected to see a continuous upsurge of cases and deaths in coming weeks. MAE describing the forecast error for Maharashtra for past 30 days is as low as 1.86% which can be attributed to the extensive testing and ...
Context 2
... in Mumbai, is of particular concern as it is considered to be one of Asia's largest slums with an area of just over 2.1 square kilometres and a population of about 1,000,000. Therefore, the rules of social distancing are difficult to observe there. Further, Mumbai has shortage of ICU beds and dedicated COVID-19 hospitals as well. As shown in Fig. 7, Maharashtra is expected to see a continuous upsurge of cases and deaths in coming weeks. MAE describing the forecast error for Maharashtra for past 30 days is as low as 1.86% which can be attributed to the extensive testing and ...
Similar publications
One of the aspects in which the COVID-19 pandemic impacted the most was education. Teachers and students had to face a new reality for which they were not prepared adapting in an improvised way new methods and strategies to teach and to learn. Within virtual education, exams reduced in some cases to multiple choice tests while others tried to mimic...
This article presented a report on an ongoing research project that integrates culture, religion and the environment. Dialectically, the Bible guides this study in the same way that this project responded to the water crisis and the environmental realities obtaining in the real world today. The paper problematizes the lack of direct citizen action...
In discussing the limits of human knowledge, physical objects are, in principle, not identical to the apparent ones. This is because the latter depend also on the observer. Virtual reality (VR) models are environments that address the senses and therefore belong to the apparent space. Therefore, they seem to prevent an epistemic justification. Base...
Pandemics that occurred in the past strongly marked all areas and spheres that makeup society. This gave rise to important and revolutionary events with repercussions that continue to this day. The COVID-19 pandemic was no different in this regard, especially as it emerged in an extremely globalized age. Therefore, the objective of this critical re...
This working paper presents the first findings of a survey on the impact of COVID-19 on workplace practices among employers in EU Member States.
The survey was developed, organised and fielded at end-2020 by Eurofound and Cedefop. It explores the impact of the health crisis on companies in the EU-27 and sheds light on how employers reacted to it....
Citations
... Thirty-seven studies forecasted unknown components in epidemiological models based on supervised learning frameworks, where AI models (primarily RNNs) were trained on synthetic data generated by epidemiological models or historical component values . Two studies did not specify their component learning frameworks 132,133 . ...
Integrating prior epidemiological knowledge embedded within mechanistic models with the data-mining capabilities of artificial intelligence (AI) offers transformative potential for epidemiological modeling. While the fusion of AI and traditional mechanistic approaches is rapidly advancing, efforts remain fragmented. This scoping review provides a comprehensive overview of emerging integrated models applied across the spectrum of infectious diseases. Through systematic search strategies, we identified 245 eligible studies from 15,460 records. Our review highlights the practical value of integrated models, including advances in disease forecasting, model parameterization, and calibration. However, key research gaps remain. These include the need for better incorporation of realistic decision-making considerations, expanded exploration of diverse datasets, and further investigation into biological and socio-behavioral mechanisms. Addressing these gaps will unlock the synergistic potential of AI and mechanistic modeling to enhance understanding of disease dynamics and support more effective public health planning and response.
... The incremental learning of the data domain refers either to tasks change defined by external supervision, human, or machine, as in the case of labels for classification, or to tasks that are drifting "naturally" from one to another because of changes in the environment that are not always identifiable (apparent drift vs. real drift). The fact is that most research works on continual learning for time series are considering task management based on an arbitrary separation of the dataset into fixed-size subsets, each one corresponding to a new task [24,27,29,[39][40][41][42][43][44][45][46]. These approaches are called Task-based approaches. ...
... An online incremental learning technique was performed along with the ANN model. They forecasted the future behavior of COVID-19 disease for the coming 30 days [46]. Christie et al. compared three forecasting methods, including ARIMA, single exponential smoothing (SES), and double exponential smoothing (DES) using the MAPE, and RMSE measures. ...
Background
There are various forecasting algorithms available for univariate time series, ranging from simple to sophisticated and computational. In practice, selecting the most appropriate algorithm can be difficult, because there are too many algorithms. Although expert knowledge is required to make an informed decision, sometimes it is not feasible due to the lack of such resources as time, money, and manpower.
Methods
In this study, we used coronavirus disease 2019 (COVID-19) data, including the absolute numbers of confirmed, death and recovered cases per day in 187 countries from February 20, 2020, to May 25, 2021. Two popular forecasting models, including Auto-Regressive Integrated Moving Average (ARIMA) and exponential smoothing state-space model with Trigonometric seasonality, Box-Cox transformation, ARMA errors, Trend, and Seasonal components (TBATS) were used to forecast the data. Moreover, the data were evaluated by the root mean squared error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE), and symmetric mean absolute percentage error (SMAPE) criteria to label time series. The various characteristics of each time series based on the univariate time series structure were extracted as meta-features. After that, three machine-learning classification algorithms, including support vector machine (SVM), decision tree (DT), random forest (RF), and artificial neural network (ANN) were used as meta-learners to recommend an appropriate forecasting model.
Results
The finding of the study showed that the DT model had a better performance in the classification of time series. The accuracy of DT in the training and testing phases was 87.50% and 82.50%, respectively. The sensitivity of the DT algorithm in the training phase was 86.58% and its specificity was 88.46%. Moreover, the sensitivity and specificity of the DT algorithm in the testing phase were 73.33% and 88%, respectively.
Conclusion
In general, the meta-learning approach was able to predict the appropriate forecasting model (ARIMA and TBATS) based on some time series features. Considering some characteristics of the desired COVID-19 time series, the ARIMA or TBATS forecasting model might be recommended to forecast the death, confirmed, and recovered trend cases of COVID-19 by the DT model.
... Farhan Mohammad Khan et al. [6] have used three different techniques such as Decision Tree, Support Vector Machine, Gaussian Process Regression algorithm to project the criticality of COVID-19 transmission in India using GIS and machine learning methods. ...
... Abdullha and Abujar [61] used LR and KNN in their model to analyze the data and predict the situation in the future and found KNN obtained higher accuracy of 99%. Farooq and Bazaz [62] used ANN to model and forecast the COVID-19 virus in the most affected parts of India and the model is intellectually able to forecast the spread of the disease. Jain et al. [31] applied ML models and Ensemble learning techniques to predict the SARS-CoV, and CoV-2 using ML models and ensemble techniques with a B-cell dataset, the higher accuracy is obtained by ensemble learning techniques. ...
The word disease is a common word and there are many diseases like heart disease, diabetes, breast cancer, COVID-19, and kidney disease that threaten humans. Data-mining methods are proving to be increasingly beneficial in the present day, especially in the field of medical applications; through the use of machine-learning methods, that are used to extract valuable information from healthcare data, which can then be used to predict and treat diseases early, reducing the risk of human life. Machine-learning techniques are useful especially in the field of health care in extracting information from healthcare data. These data are very much helpful in predicting the disease early and treating the patients to reduce the risk of human life. For classification and decision-making, data mining is very much suitable. In this paper, a comprehensive study on several diseases and diverse machine-learning approaches that are functional to predict those diseases and also the different datasets used in prediction and making decisions are discussed in detail. The drawbacks of the models from various research papers have been observed and reveal countless computational intelligence approaches. Naïve Bayes, logistic regression (LR), SVM, and random forest are able to produce the best accuracy. With further optimization algorithms like genetic algorithm, particle swarm optimization, and ant colony optimization combined with machine learning, better performance can be achieved in terms of accuracy, specificity, precision, recall, and specificity.
... Machine learning (ML) has proven itself over time to be an important aspect to big data analytics (Athmaja, Hanumanthappa, and Kavitha 2017). Recently, ML has been applied to many real-time tasks such as COVID-19 modeling and prediction (Farooq and Bazaz 2021;Lalmuanawma, Hussain, and Chhakchhuak 2020), prognostics of machinery (Wen et al. 2022;Cummins et al. 2021;Adhikari, Rao, and Buderath 2018), and more. Most of these applications follow the same offline learning paradigm where a model is trained and deployed. ...
Machine learning has shown to be a crucial part of big data analytics; however, it lacks when the data is continuously streaming in from the system and changing too much from the original training data. Online learning is machine learning for streaming data that arrives in a sequential order where the model updates after every data point. While machine learning relies on well-established libraries such as PyTorch and Keras, the libraries for online learning are less well known, but they are here to serve similar purposes of reproducibility and reducing the time from research to production. Here, we compare different libraries for online learning research, specifically supervised learning. We compare them on the axes of developmental experience and benchmark testing as researchers. Our comparison as developers takes maintenance, documentation, and offerings of state-of-the-art algorithms into account. As this is not necessarily free of bias, we also use benchmarks known to online learning to gather power usage, RAM usage, speed, and accuracy of these libraries to get an objective view. Our findings show that Avalanche and River, including River-torch, are among the best libraries in terms of performance and applicability to the research in supervised online learning.
... eir experiments showed that LSTM outperformed ARIMA. Furthermore, Farooq and Bazaz, [42] have developed a method that forecasts the pandemic in India using the artificial neural network (ANN). e model used the online incremental learning technique, in which its parameters were adapted intelligently to a new dataset, and was able to forecast the cases 30 days ahead in five badly-affected Indian states. ...
The spread of COVID-19 has affected more than 200 countries and has caused serious public health concerns. The infected cases are on the increase despite the effectiveness of the vaccines. An efficient and quick surveillance system for COVID-19 can help healthcare decision-makers to contain the virus spread. In this study, we developed a novel framework using machine learning (ML) models capable of detecting COVID-19 accurately at an early stage. To estimate the risks, many models use social networking sites (SNSs) in tracking the disease outbreak. Twitter is one of the SNSs that is widely used to create an efficient resource for disease real-time analysis and can provide an early warning for health officials. We introduced a pipeline framework of outbreak prediction that incorporates a first-step hybrid method of word embedding for tweet classification. In the second step, we considered the classified tweets with external features such as vaccine rate associated with infected cases passed to machine learning algorithms for daily predictions. Thus, we applied different machine learning models such as the SVM, RF, and LR for classification and the LSTM, Prophet, and SVR for prediction. For the hybrid word embedding techniques, we applied TF-IDF, FastText, and Glove and a combination of the three features to enhance the classification. Furthermore, to improve the forecast performance, we incorporated vaccine data as input together with tweets and confirmed cases. The models’ performance is more than 80% accurate, which shows the reliability of the proposed study.
... By considering a dataset of seventy-seven days, they determined the COVID-19 infected, recovery, and death cases for the next seven days. In related work, authors of [49] considered the ARIMA and FTS models to forecast the COVID-19 infection cases in Egypt, South Africa, and Algeria wherein the ARIMA method harmonized with the data trajectory, whereas, in their work, the authors of [50] developed the ANN-based prediction approach to predict the COVID-19 total, active, and death cases in different states of India. ...
The second wave of the COVID-19 pandemic outburst triggered enormously all over India. This ill-fated and fatal brawl affected millions of Indian citizens, with many active and infected Indians struggling to recover from this deadly disease to date, leading to a grief situation. The present situation warrants developing a robust and sound forecasting model to evaluate the adversities of the epidemic with reasonable accuracy to assist officials in curbing this hazard. Consequently, we employed Auto-ARIMA, Auto-ETS, Auto-MLP, Auto-ELM, AM, MLP and proposed ELM methods for assessing accumulative infected COVID-19 individuals by the end of July 2021. We made 90 days of advanced forecasting, i.e., up to 24 July 2021, for the number of cumulative infected COVID-19 cases of India using all seven methods in 15 days’ intervals. We fine-tuned the hyper-parameters to enhance the prediction performance of these models and observed that the proposed ELM model offers satisfactory accuracy with MAPE of 5.01, and it rendered better accuracy than the other six models. To comprehend the dataset's nature, five features are extracted. The resulting feature values encouraged further investigation of the models for an updated dataset, where the proposed model provides encouraging results.
... Similarly, the authors in [86] compared the performance of linear regression, MLP, and vector autoregression in forecasting the number of cases in India. Although MLP has not been utilized successfully as a standalone model, it was shown to be helpful in estimating the optimal coefficients of the SEIR model [25]. In [93], the authors use a SEIRbased teacher simulation model to obtain projection sequences which are used together with the original sequences to train the student MLP model to make accurate forecasts. ...
The Covid-19 pandemic has galvanized scientists to apply machine learning methods to help combat the crisis. Despite the significant amount of research there exists no comprehensive survey devoted specifically to examining deep learning methods for Covid-19 forecasting. In this paper, we fill the gap in the literature by reviewing and analyzing the current studies that use deep learning for Covid-19 forecasting. In our review, all published papers and preprints, discoverable through Google Scholar, for the period from Apr 1, 2020 to Feb 20, 2022 which describe deep learning approaches to forecasting Covid-19 were considered. Our search identified 152 studies, of which 53 passed the initial quality screening and were included in our survey. We propose a model-based taxonomy to categorize the literature. We describe each model and highlight its performance. Finally, the deficiencies of the existing approaches are identified and the necessary improvements for future research are elucidated. The study provides a gateway for researchers who are interested in forecasting Covid-19 using deep learning.
... They found that the ensembling system is outperformed by (DLM-system) Deep learning models for four countries [12]. There are further studies that focus on the application of deep learning models to the prediction of SARSC OV2 [36][37][38][39]. ...
Artificial intelligence (AI), machine learning (ML), and deep learning (DL) are just a few of the technologies that can help healthcare organizations support real-time decision-making to control the spread of a pandemic. This study aims to investigate hyper-parameter tuning for Long Short-Term Memory (LSTM) to forecast SARSCoV2 daily infection cases in the Russian Federation by selecting the best loss function, activation function, number of epochs, number of neurons in a cell, and optimizer to minimize the error in addition to a good fit for the model on both the training and validation sets. In the meanwhile, we use LSTM, LSTMs (stacked LSTM), bidirectional LSTM, BILSTMs (stacked BILSTM), convolution neuron networks (Conv), ConvLSTMs and other forecasting models to forecast the daily SARSCoV2 infection cases one by one. We analyzed and compared the results obtained by these models. We discovered that BiLSTM can efficiently extract features from the data, which are the items from the previous day. With the extracted feature data, we used this model to forecast daily infection cases. We used BiLSTM to forecast SARSCoV2 spreading in the Russian Federation for one month. Hence, the Bidirectional LSTM can accurately predict daily SARSCoV2 infection cases in Russia, according to experiments.
