# Optimal Neural Network Model for Short-Term Prediction of Confirmed Cases in the COVID-19 Pandemic

^{1}

^{2}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Materials and Methods

#### 2.1. Study Area and Data

#### 2.2. Forecasting Methodogy

_{i}, y

_{i}), i = 1, 2,..., N (N is the number of available samples) is a collection of observables of an unknown function ŷ = f(t) collected at regular time intervals ∆t, where t

_{i+1}= t

_{i}+ ∆t. Data collected in the past are utilized to estimate the future trends of the observed variable during the forecasting process. In order to enable one-step-ahead predictions, it is necessary to find the optimal mathematical form of the function f(t) that can correctly carry out the following transformation:

_{i+1}) = ŷ(I + 1) + ε,

#### 2.3. Prediction Accuracy Measures

_{i}and ŷ

_{i}are the obtained and expected value of the forecast, respectively.

^{2}: This coefficient can range between 0 and 1 and shows how well a forecasting model predicts the outcome. It is a measure of the goodness of fit. Its higher value corresponds to a better prediction for a model. It can be calculated as (9):

## 3. Results and Discussion

^{2}of the system’s forecasts for all six analyzed countries. This is shown in Table 3.

^{2}as a measure of goodness of fit ranged from 0.67 for Canada to 0.99 for Chile. The corresponding performance measures in the case of the suggested forecasting methodology performed better than ELM modeling, considering all analyzed types of forecasting errors, as shown in Table 3. For example, the calculated R

^{2}values for our methodology ranged from 0.9896 for Canada to 0.9990 for Malaysia.

## 4. Conclusions and Future Work

## Author Contributions

## Funding

## Data Availability Statement

## Conflicts of Interest

## References

- Wang, C.; Horby, P.W.; Hayden, F.G.; Gao, G.F. A novel corona virus outbreak of global health concern. Lancet
**2020**, 395, 470–473. [Google Scholar] [CrossRef] [Green Version] - WHO Coronavirus (COVID-19) Dashboard. Available online: https://covid19.who.int/ (accessed on 15 September 2022).
- Nasserie, T.; Hittle, M.; Goodman, S.N. Assessment of the Frequency and Variety of Persistent Symptoms Among Patients with COVID-19 A Systematic Review. JAMA Netw. Open
**2021**, 4, e2111417. [Google Scholar] [CrossRef] - Wang, P.; Lu, J.A.; Jin, Y.; Zhu, M.; Wang, L.; Chen, S. Statistical and network analysis of 1212 COVID-19 patients in Henan, China. Int. J. Infect. Dis.
**2020**, 95, 391–398. [Google Scholar] - Gao, Y.D.; Ding, M.; Dong, X.; Zhang, J.J.; Kursat Azkur, A.; Azkur, D.; Gan, H.; Sun, Y.L.; Fu, W.; Li, W.; et al. Risk factors for severe and critically ill COVID-19 patients: A review. Allergy
**2021**, 76, 428–455. [Google Scholar] [CrossRef] - Turk, M.A.; Landes, S.D.; Formica, M.K.; Goss, K.D. Intellectual and developmental disability and COVID-19 case-fatality trends: TriNetXanalysis. Disabil. Health J.
**2020**, 13, 100942. [Google Scholar] [CrossRef] [PubMed] - Ceylan, Z. Short-term prediction of COVID-19 spread using grey rolling model optimized by particle swarm optimization. Appl. Soft Comput.
**2021**, 109, 107592. [Google Scholar] [CrossRef] [PubMed] - Gomez-Cravioto, D.A.; Diaz-Ramos, R.E.; Cantu-Ortiz, F.J.; Ceballos, H.G. Data Analysis and Forecasting of the COVID-19 Spread: A Comparison of Recurrent Neural Networks and Time Series Models. Cogn. Comput.
**2021**, 1–12. [Google Scholar] [CrossRef] - Pavlyutin, M.; Samoyavcheva, M.; Kochkarov, R.; Pleshakova, E.; Korchagin, S.; Gataullin, T.; Nikitin, P.; Hidirova, M. COVID-19 Spread Forecasting, Mathematical Methods vs. Machine Learning, Moscow Case. Mathematics
**2022**, 10, 195. [Google Scholar] [CrossRef] - Frausto-Solis, J.; Hernández-González, L.J.; González-Barbosa, J.J.; Sánchez-Hernández, J.P.; Román-Rangel, E. Convolutional Neural Network–Component Transformation (CNN–CT) for Confirmed COVID-19 Cases. Math. Comput. Appl.
**2021**, 26, 29. [Google Scholar] [CrossRef] - Borghi, P.H.; Zakordonets, O.; Teixeira, J.P. A COVID-19 time series forecasting model based on MLP ANN. Procedia Comput. Sci.
**2021**, 181, 940–947. [Google Scholar] [CrossRef] [PubMed] - Milić, M.; Milojković, J.; Marković, I.; Nikolić, P. Concurrent, Performance-Based Methodology for Increasing the Accuracy and Certainty of Short-Term Neural Prediction Systems. Comput. Intell. Neurosci.
**2019**, 2019, 9323482. [Google Scholar] [CrossRef] - Shakhovska, N.; Melnykova, N.; Chopiyak, V.; Gregus, M. An ensemble method for medical insurance costs prediction task. Comput. Mater. Contin.
**2022**, 70, 3969–3984. [Google Scholar] [CrossRef] - Tkachenko, R.; Izonin, I. Model and principles for the implementation of neural-like structures based of geometric data transformations. In International Conference on Computer Science, Engineering and Education Applications; Springer: Berlin/Heidelberg, Germany, 2018; pp. 578–587. [Google Scholar]
- Tkavhenko, R.; Izonin, I.; Vitynskyi, P.; Lotoshunska, N.; Pavlyuk, O. Development of the Non-Iterative Supervised Learning Predictor Based on the Ito Decomposition and SGTM Neural-Like Structure for Managing Medical Insurance Costs. Data
**2018**, 3, 46. [Google Scholar] [CrossRef] [Green Version] - Ahuja, S.; Shelke, N.A.; Singh, P.K. A deep learning framework using CNN and stacked Bi-GRU for COVID-19 predictions in India. Signal Image Video Process
**2022**, 16, 579–586. [Google Scholar] [CrossRef] - Al-qaness, M.A.; Fan, H.; Ewees, A.A.; Yousri, D.; Elaziz, M.A. Improved ANFIS model for forecasting Wuhan City Air Quality and analysis COVID-19 lockdown impacts on air quality. Environ. Res.
**2021**, 194, 110607. [Google Scholar] [CrossRef] - Polishchuk, E.A. The Analysis of the Selection Criteria of the Optimal Model of the Dynamics in the Case of Extrapolative Forecasting for Short Time Series. In Proceedings of the International Scientific Conference, Far East Con, Online, 17 March 2020; Atlantis Press: Amsterdam, The Netherlands, 2020. [Google Scholar]
- Hernandez-Matamoros, A.; Fujita, H.; Hayashi, T.; Perez-Meana, H. Forecasting of COVID19 per regions using ARIMA models and polynomial functions. Appl. Soft Comput.
**2020**, 96, 106610. [Google Scholar] [CrossRef] [PubMed] - Miyama, T.; Jung, S.M.; Hayashi, K.; Anzai, A.; Kinoshita, R.; Kobayashi, T.; Linton, N.M.; Suzuki, A.; Yang, Y.; Yuan, B.; et al. Phenomenological and mechanistic models for predicting early transmission data of COVID-19. Math. Biosci. Eng.
**2022**, 19, 2043–2055. [Google Scholar] [CrossRef] - Friston, K.J.; Parr, T.; Zeidman, P.; Razi, A.; Flandin, G.; Daunizeau, J.; Hulme, O.J.; Billig, A.J.; Litvak, V.; Moran, R.J.; et al. Dynamic causal modelling of COVID-19. Welcome Open Res.
**2020**, 5, 04463. [Google Scholar] - Tealab, A.; Hefny, H.; Badr, A. Forecasting of nonlinear time series using ANN. Future Comput. Inform. J.
**2017**, 2, 39–47. [Google Scholar] - Maier, H.R.; Dandy, G.C. Neural Network Based Modelling of Environmental Variables: A Systematic Approach. Math. Comput. Model.
**2001**, 33, 669–682. [Google Scholar] [CrossRef] - Lu, Y.; Mei, G.; Piccialli, F.A. Deep Learning Approach for Predicting Two-Dimensional Soil Consolidation Using Physics-Informed Neural Networks (PINN). Comput. Eng. Financ. Sci.
**2022**, 10, 05710. [Google Scholar] [CrossRef] - Qin, Y.; Zhao, M.; Lin, Q.; Li, X.; Ji, J. Data-Driven Building Energy Consumption Prediction Model Based on VMD-SA-DBN. Mathematics
**2022**, 10, 3058. [Google Scholar] [CrossRef] - He, L.; Kong, D.; Lei, Z. Research on Vibration Propagation Law and Dynamic Effect of Bench Blasting. Mathematics
**2022**, 10, 2951. [Google Scholar] [CrossRef] - Bagnasco, A.; Siri, A.; Aleo, G.; Rocco, G.; Sasso, L. Applying artificial neural networks to predict communication risks in the emergency department. J. Adv. Nurs.
**2015**, 71, 2293–2304. [Google Scholar] [CrossRef] - Shahid, N.; Rappon, T.; Berta, W. Applications of artificial neural networks in health care organizational decision-making: A scoping review. PLoS ONE
**2019**, 14, e0212356. [Google Scholar] [CrossRef] - World Map. Available online: https://www.mapchart.net/world.html (accessed on 15 September 2022).
- Trepanowski, R.; Drazkowski, D. Cross-National Comparison of Religion as a Predictor of COVID-19 Vaccination Rates. J. Relig. Health
**2022**, 61, 2198–2211. [Google Scholar] [CrossRef] - Tang, J.W.; Caniza, M.A.; Dinn, M.; Dwyer, D.E.; Heraud, J.M.; Jennings, L.C.; Kok, J.; Kwok, K.O.; Li, Y.; Loh, T.P.; et al. An exploration of the political, social, economic and cultural factors affecting how different global regions initially reacted to the COVID-19 pandemic. Interface Focus
**2022**, 12, 20210079. [Google Scholar] [CrossRef] - Wang, Y.; Yan, Z.; Wang, D.; Yang, M.; Li, Z.; Gong, X.; Wu, D.; Zhai, L.; Zhang, W.; Wang, Y. Prediction and analysis of COVID-19 daily new cases and cumulative cases: Times series forecasting and machine learning models. BMC Infect. Dis.
**2022**, 22, 495. [Google Scholar] [CrossRef] - Abu-Abdoun, D.I.; Al-Shihabi, S. Weather Conditions and COVID-19 Cases: Insights from the GCC Countries. Intell. Syst. Appl.
**2022**, 15, 200093. [Google Scholar] [CrossRef] - Allain-Dupré, D.; Chatry, I.; Michalun, V.; Moisio, A. OECD Policy Responses Coronavirus (COVID-19), The Territorial Impact of COVID-19: Managing the Crisis across Levels of Government; OECD: Paris, France, 2020.
- Available online: https://worldometer.info/coronavirus (accessed on 15 September 2022).
- Jones, J.S.; Goldring, J. Exploratory and Descriptive Statistics; SAGE Publications Ltd.: London, UK, 2022. [Google Scholar]
- Jha, G.K.; Sinha, K. Time-delay neural networks for time series prediction: An application to the monthly wholesale price of oilseeds in India. Neural Comput. Appl.
**2014**, 24, 563–571. [Google Scholar] [CrossRef] - Milojković, J.; Milić, M.; Litovski, V. ANN model for one day ahead Covid-19 prediction. In Proceedings of the Conference IcETRAN, Novi Pazar, Serbia, 6–8 June 2022; pp. 296–300. [Google Scholar]
- Masters, T. Practical Neural Network Recipes in C++; Elsevier: Amsterdam, The Netherlands, 1993. [Google Scholar]
- Yuehjen, E.S.; Lin, S.C. Using a Time Delay Neural Network Approach to Diagnose the Out-of-Control Signals for a Multivariate Normal Process with Variance Shifts. Mathematics
**2019**, 7, 959. [Google Scholar] - Zhang, G.; Eddy Patuwo, B.; Hu, Y.M. Forecasting with artificial neural networks. Int. J. Forecast.
**1998**, 1, 35–62. [Google Scholar] [CrossRef] - Milojković, J.; Litovski, V. Dynamic Short-Term Forecasting of Electricity Load Using Feed-Forward ANNs. Eng. Intell. Syst. Electr. Eng. Commun.
**2009**, 17(1), 39–48. [Google Scholar] - Milojković, J.; Litovski, V. One step ahead prediction in electronics based on limited information. In Proceedings of the Conference ETRAN, Vrnjačka Banja, Serbia, 15–18 June 2019; p. EL1.7. [Google Scholar]
- Milojković, J.; Litovski, V. Short-Term Forecasting of Electricity Load Using Recurrent ANNs. In Proceedings of the Electronics, Banja Luka, Bosnia and Hercegovina, 14 June 2010; pp. 45–49. [Google Scholar]
- Meza, J.C. Steepest descent. In Wiley Interdisciplinary Reviews: Computational Statistics 2.6 2010; Willey: Hoboken, NJ, USA, 2010; pp. 719–722. [Google Scholar]
- Drummond, L.M.G.; Svaiter, B.F. A steepest descent method for vector optimization. J. Comput. Appl. Math.
**2005**, 175, 395–414. [Google Scholar] [CrossRef] [Green Version] - Fletcher, R. A limited memory steepest descent method. Math. Program.
**2012**, 135, 413–436. [Google Scholar] [CrossRef] - Narkhede, M.V.; Bartakke, P.P.; Sutaone, M.S. A review on weight initialization strategies for neural networks. Artif. Intell. Rev.
**2022**, 55, 291–322. [Google Scholar] [CrossRef] - Yam, J.Y.F.; Chow, T.W.S. A weight initialization method for improving training speed in feedforward neural network. Neurocomputing
**2000**, 30, 219–232. [Google Scholar] [CrossRef] - Denoeux, T.; Lengelle, R. Initializing backpropagation networks with prototypes. Neural Netw.
**1993**, 6, 351–363. [Google Scholar] [CrossRef] - Raschka, S.; Mirjalili, V. Python Machine Learning; Packt Publishing: Birmingham, UK, 2019. [Google Scholar]
- Milić, M.; Milojković, J.; Jeremić, M. A deep learning approach for hydrological time-series prediction with ELM model. In Proceedings of the Small Systems Simulation Symposium, Niš, Serbia, 28 February–2 March 2022; Atlantis Press: Amsterdam, The Netherland, 2022; pp. 61–65. [Google Scholar]
- Das, A.K.; Mishra, D.; Das, K.; Mallick, P.K.; Kumar, S.; Zymbler, M.; El-Sayed, H. Prophesying the Short-Term Dynamics of the Crude Oil Future Price by Adopting the Survival of the Fittest Principle of Improved Grey Optimization and Extreme Learning Machine. Mathematics
**2022**, 10, 1121. [Google Scholar] [CrossRef] - Nguyen, D.-T.; Ho, J.-R.; Tung, P.-C.; Lin, C.-K. Prediction of Kerf Width in Laser Cutting of Thin Non-Oriented Electrical Steel Sheets Using Convolutional Neural Network. Mathematics
**2021**, 9, 2261. [Google Scholar] [CrossRef] - Díaz-Narváez, V.; San-Martín-Roldán, D.; Calzadilla-Núñez, A.; San-Martín-Roldán, P.; Parody-Muñoz, A.; Robledo-Veloso, G. Which curve provides the best explanation of the growth in confirmed COVID-19 cases in Chile? Rev. Lat.-Am. Enferm.
**2020**, 28, 1–9. [Google Scholar] [CrossRef] [PubMed] - Medeiros, M.C.; Street, A.; Valladao, D.; Vasconcelos, G.; Zilberman, E. Short-term COVID-19 forecast for latecomers. Int. J. Forecast. B
**2022**, 38, 467–488. [Google Scholar] [CrossRef] [PubMed] - Bekker, R.; Broek, M.; Koole, G. Modeling COVID-19 hospital admissions and occupancy in the Netherlands. Eur. J. Oper. Res.
**2023**, 304, 207–218. [Google Scholar] [CrossRef] - Ahterberg, M.A.; Prasse, B.; Ma, L.; Trajanovski, S.; Kitsak, M.; Mieghem, P.V. Comparing the accuracy of several network-based COVID-19 prediction algorithms. Int. J. Forecast.
**2022**, 38, 489–504. [Google Scholar] [CrossRef] - Chimmula, V.K.R.; Zhang, L. Time series forecasting of COVID-19 transmission in Canada using LSTM networks. Chaos Solitons Fractals
**2020**, 135, 109864. [Google Scholar] [CrossRef] - Zhang, J.; Pathak, H.S.; Snowdon, A.; Greiner, R. Learning models for forecasting hospital resource utilization for COVID-19 patients in Canada. Sci. Rep.
**2022**, 12, 8751. [Google Scholar] [CrossRef] - Ahmar, A.S.; del Val, E.B. SutteARIMA: Short-term forecasting method, a case: COVID-19 and stock market in Spain. Sci. Total Environ.
**2020**, 729, 138883. [Google Scholar] [CrossRef] - Castillo, O.; Melin, P. Forecasting of COVID-19 time series for countries in the world based on a hybrid approach combining the fractal dimension and fuzzy logic. Chaos Solitons Fractals
**2020**, 140, 110242. [Google Scholar] [CrossRef] - Appadu, A.A.; Kelil, A.S.; Tijani, Y.O. Comparison of some forecasting methods for COVID-19. Alex. Eng. J.
**2021**, 60, 1565–1589. [Google Scholar] [CrossRef] - Perc, M.; Gorišek Miksić, N.; Slavinec, M.; Stožer, A. Forecasting COVID-19. Front. Phys.
**2020**, 8, 00127. [Google Scholar] [CrossRef] [Green Version] - Alsayed, A.; Sadir, H.; Kamil, R.; Sari, H. Prediction of Epidemic Peak and Infected Cases for COVID-19 Disease in Malaysia. Int. J. Environ. Res. Public Health
**2020**, 17, 4076. [Google Scholar] [CrossRef] - Salim, N.; Chan, W.H.; Mansor, S.; Bazin, N.E.; Amaran, S.; Faudzi, A.A.; Zainal, A.; Huspi, S.H.; Hooi, E.K.; Shithil, S.M. COVID-19 epidemic in Malaysia: Impact of lockdown on infection dynamics. medRxiv
**2020**. [Google Scholar] [CrossRef] [Green Version] - Shetty, R.P.; Pai, P.S. Forecasting of COVID 19 cases in Karnataka state using artificial neural network (ANN). J. Inst. Eng. Ser. B
**2021**, 102, 1201–1211. [Google Scholar] [CrossRef] - Wieczorek, M.; Siłka, J.; Woźniak, M. Neural network powered COVID-19 spread forecasting model. Chaos Solitons Fractals
**2020**, 140, 110203. [Google Scholar] [CrossRef] [PubMed] - Namasudra, S.; Dhamodharavadhani, S.; Rathipriya, R. Nonlinear neural network based forecasting model for predicting COVID-19 cases. Neur. Proc. Lett.
**2021**, 1–21. [Google Scholar] [CrossRef] - Alazab, M.; Awajan, A.; Mesleh, A.; Abraham, A.; Jatana, V.; Alhyari, S. COVID-19 prediction and detection using deep learning. Int. J. Comp. Inf. Syst. Ind. Manag. Appl.
**2020**, 12, 168–181. [Google Scholar] - Rauf, H.T.; Lali, M.; Khan, M.A.; Kadry, S.; Alolaiyan, H.; Razaq, A.; Irfan, R. Time series forecasting of COVID-19 transmission in Asia Pacific countries using deep neural networks. Pers. Ubiquitous Comp.
**2021**, 1–8. [Google Scholar] [CrossRef]

**Figure 5.**Prediction diagrams of COVID-19 daily confirmed infection cases obtained for the year 2021, versus the real daily number of infection cases for (

**a**) Chile, (

**b**) the Netherlands, (

**c**) Canada, (

**d**) Spain, (

**e**) South Africa, and (

**f**) Malaysia.

**Figure 6.**Relative forecasting accuracy improvement from individual TDNNC blocks with the number of neurons in the hidden layer varying from three to ten, to the constructed optimal forecasting system, for six world countries.

**Table 1.**Descriptive statistics of the COVID-19 daily spread datasets for Chile, the Netherlands, Canada, Spain, South Africa, and Malaysia for 2021.

Descriptive Statistics | Chile | Netherlands | Canada | Spain | South Africa | Malaysia |
---|---|---|---|---|---|---|

Number of scores | 365 | 365 | 365 | 365 | 365 | 365 |

Mean | 3290.54 | 6348.32 | 4147.86 | 11,340.40 | 6575.88 | 7249.05 |

Median | 2903 | 4758.5 | 3230 | 7020 | 3514 | 5343 |

25th Percentile | 1238.5 | 7493.75 | 2343.5 | 4192 | 1195.25 | 3086 |

75th Percentile | 5360.75 | 17,959.25 | 5079 | 15,095.5 | 11488 | 8636.2.5 |

Interquartile Range | 4122.25 | 4969.25 | 2735.5 | 10,903.5 | 10,292.75 | 5550.25 |

Minimum | 421 | 588 | 377 | 1464 | 259 | 1211 |

Maximum | 7321 | 22,471 | 32370 | 88,040 | 23437 | 21,808 |

Range | 6900 | 32,400 | 31993 | 86,576 | 23178 | 20,597 |

Variance | 4,951,854.29 | 27,420,731.59 | 13,255,497.17 | 134,112,484.02 | 40,227,254.55 | 35,589,921.39 |

Standard Deviation | 2225.28 | 5236.48 | 3640.81 | 11,580.69 | 6342.50 | 5965.73 |

Skew | 0.38 | 1.61 | 3.53 | 2.75 | 0.79 | 1.22 |

Kurtosis | −1.24 | 2.08 | 19.53 | 11.44 | −0.66 | 0.19 |

**Table 2.**Averaged relative error of the entire forecasting system in [%] with the relative

**accuracy increases**achieved using the optimal TDNNC based forecasting methodology in comparison with the individual TDNNCs.

Err. of the System | System vs. 3 h. n. TDNNC | System vs. 4 h. n. TDNNC | System vs. 5 h. n. TDNNC | System vs. 6 h. n. TDNNC | System vs. 7 h. n. TDNNC | System vs. 8 h. n. TDNNC | System vs. 9 h. n. TDNNC | System vs. 10 h. n. TDNNC | |
---|---|---|---|---|---|---|---|---|---|

Chile | 1.90 | 4.52% | 5.00% | 4.52% | 4.52% | 9.52% | 7.77% | 6.86% | 7.77% |

Netherl. | 2.75 | 8.94% | 10.42% | 7.41% | 4.52% | 8.33% | 4.84% | 2.13% | 1.45% |

Canada | 3.75 | 4.58% | 3.35% | 5.06% | 10.29% | 2.09% | 6.01% | 6.25% | 2.85% |

Spain | 3.16 | 10.73% | 12.46% | 9.71% | 5.39% | 7.6% | 11.23% | 10.73% | 11.23% |

S. Africa | 3.24 | 5.54% | 4.98% | 4.71% | 7.16% | 5.81% | 4.42% | 2.11% | 2.99% |

Malaysia | 2.23 | 18.91% | 32.42% | 16.48% | 3.04% | 5.11% | 1.76% | 1.76% | 0.45% |

**Table 3.**Performance measures (RMSE, MAPE and R

^{2}) for the optimal forecasting system, and for the Extreme Learning Machine forecasting methodology for different countries.

Country | |||||||
---|---|---|---|---|---|---|---|

Accuracy Measure | Chile | Netherlands | Canada | Spain | S. Africa | Malaysia | |

Optimal neural forecasting system | RMSE | 88.67 | 253.80 | 393.74 | 786.68 | 429.43 | 192.30 |

MAPE [%] | 1.90 | 2.75 | 3.75 | 3.15 | 3.24 | 2.23 | |

R^{2} | 0.9984 | 0.9977 | 0.9896 | 0.9955 | 0.9955 | 0.9990 | |

Extreme Learning Machine | RMSE | 265.46 | 1110.15 | 3134.49 | 7380.34 | 2867.22 | 1283.14 |

MAPE [%] | 4.66 | 7.8 | 22.94 | 16.68 | 10.77 | 5.69 | |

R^{2} | 0.99 | 0.96 | 0.67 | 0.79 | 0.69 | 0.96 |

**Table 4.**Systematization of the current state-of-the-art COVID-19 related forecasting solutions, and their reported performance measures for six analyzed countries.

Country | Methodology | Performance Measure | Ref. | ||
---|---|---|---|---|---|

RMSE | Prediction Error [%] | MAPE [%] | |||

Chile | Different exponential smoothing algorithms | 1.06–1.74 | - | 7.4 | [55] |

Penalized LASSO regression model with an error correction mechanism | - | - | 0.32–5.22 | [56] | |

Netherlands | Linear programing method | 5.76–16.15 | - | 5.9–8.2 | [57] |

NIPA | - | - | 0.04 | [58] | |

Logistic function | - | - | 0.07 | [58] | |

Canada | LSTM | 34.83–51.46 | 6.6-7.33 | - | [59] |

Multiple Temporal Convolutional network | - | 6–9.47 | [60] | ||

Spain | Sutte ARIMA | - | - | 0.036 | [61] |

Fuzzy Fractal Approach | - | 3.58139 | - | [62] | |

S. Africa | Hybrid-Euler | - | 0.56-35.63 | - | [63,64] |

Malaysia | ANFIS | 46.85 | - | 1.31 | [65] |

ARIMA | - | - | 16.01 | [66] |

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Milić, M.; Milojković, J.; Jeremić, M.
Optimal Neural Network Model for Short-Term Prediction of Confirmed Cases in the COVID-19 Pandemic. *Mathematics* **2022**, *10*, 3804.
https://doi.org/10.3390/math10203804

**AMA Style**

Milić M, Milojković J, Jeremić M.
Optimal Neural Network Model for Short-Term Prediction of Confirmed Cases in the COVID-19 Pandemic. *Mathematics*. 2022; 10(20):3804.
https://doi.org/10.3390/math10203804

**Chicago/Turabian Style**

Milić, Miljana, Jelena Milojković, and Miljan Jeremić.
2022. "Optimal Neural Network Model for Short-Term Prediction of Confirmed Cases in the COVID-19 Pandemic" *Mathematics* 10, no. 20: 3804.
https://doi.org/10.3390/math10203804