|
| S. No. | Author and year | Methodology adopted for battling against COVID-19 | Findings |
|
| 1 | Pham et al. (2020) | Artificial intelligence and big data | The authors highlighted the techniques to detect, diagnose, track, and predict the spread, vaccine, and drug discovery using both artificial intelligence and big data approaches and recommended some points to be taken by the authorities. |
|
| 2 | Shahid et al. (2020) | Machine learning | The authors presented various machine learning techniques to track and forecast the disease, medical abetment, and virus detection in the human body and also defined the approach to be considered for the proper understanding of virology to combat the pandemic. |
|
| 3 | Nayak and Naik (2021) | Deep learning and machine learning | The authors used machine learning and deep learning approaches for the screening and diagnosing of COVID-19. The authors also presented various ML algorithms such as SVM, KNN, LF, and RF that can be used for dealing with COVID-19. The author proposed the research community to use optimization method and fuzzy techniques of ML for future projection of COVID-19. |
|
| 4 | (Alsunaidi et al., 2021) | Big data analytics | The authors presented various applications of data analysis for COVID-19 and provided a hierarchical structure to classify these applications into four classes: diagnosis, predicting the risk factor, decision-making for treatment to be provided, and medication. |
|
| 5 | Behnam and Jahanmahin (2021) | Descriptive and prescriptive data analytics | The authors presented a study to predict the frequency of the trend of COVID-19 in order to get the best fitting pattern for prediction of the peak and end of the pandemic in various time slots and also analyzed the rate of mortality, rate of recovery, rate of spread of infection, and weekly data about the disease. The authors concluded that the Gaussian function is best among all techniques to predict the spread of COVID-19 and other important parameters as it can presume the curve with more accuracy. |
|
| 6 | (Swapnarekha et al., 2020) | Deep learning and artificial neural network | The researchers concluded that SVM, linear regression, K means, and RF approaches can be used to solve the remedial issues of COVID-19, while CNN and deep learning can be used for the early prediction of COVID-19. The authors further stated that although these techniques are much useful in prediction and diagnosis, still, lack of data sets and medical images leads to inaccurate results of the research that further can be sorted out using bagging, stacking, and so on. |
|
| 7 | (Car et al., 2020) | Multilayer perceptron model | The authors collected the data sets, applied those data sets to the multilayer perceptron model, and reached the conclusion that it is feasible to get a high-quality model that can predict the various parameters involved in the spread of COVID-19 by taking the geographical data and time as the input data. For future work, the authors suggested using some testing techniques of various parameters (such as percentage of infected population) to check the quality of the model. |
|
| 8 | (Tuli et al., 2020) | Cloud computing and machine learning | The authors discussed how enhanced mathematical modeling in collaboration with cloud computing and machine learning help in predicting the broadening of COVID-19 at the initial stages. A model named robust Weibull model was proposed to make the statistical forecast and compared this with the Gaussian model and finally concluded that the robust Weibull model is more accurate for making predictions |
|
| 9 | Syeda et al. (2021) | CNN model in machine learning | The authors reviewed that there can be various application areas for handling the pandemic where machine learning can be effectively utilized ranging from tracing and curing the viral infection. The researchers further presented some light on machine-learning-based tools for the discovery of appropriate drug. |
|
| 10 | Latif et al. (2020) | Data science | The authors used the available data sets on COVID-19 and presented a bibliometric analysis to keep on the track spread and weakening strategies of the virus. |
|
| 11 | (Ghorui et al., 2021a) | Multicriteria decision-making | The researchers analyzed the most dominant risk factor in spreading COVID-19 using the HFS TOPSIS and fuzzy AHP approach. |
|
| 12 | (Samanlioglu & Kaya, 2020) | Fuzzy AHP technique | The authors proposed the MCDM-based hesitant F-AHP technique that can be implemented by governments, healthcare service providers, and decision-makers in order to improve the condition or prevent it from getting worse because of pandemic. |
|
| 13 | Mondal et al. (2020) | Data analytics | The authors used polynomial regression for the modeling of increment in a number of confirmed cases. The authors also proved that LR, MLP, and XGBoost approaches can be very effective for classifying the patients suffering from COVID-19. |
|
| 14 | (Łuczak & Kalinowski, 2022) | Fuzzy approach | The authors have used fuzzy C means clustering to identify the most affected states of COVID-19. |
|
| 15 | (Alotaibi & Elaraby, 2022) | Exponential fuzzy approach | The author of the paper has used a multithreshold approach using a generalized exponential entropy-based algorithm for the segmentation of CT scans for identifying COVID |
|
