TÜRKİYE'DE DOĞRULANMIŞ COVİD-19 VAKALARININ İSTİFLEME TOPLULUK MODELLER İLE ZAMAN SERİSİ TAHMİNİ

Öz

COVID-19 hemen hemen her ülkeye yayıldığı ve insanlık için ciddi bir tehdit oluşturduğu için DSÖ tarafından pandemi olarak ilan edilmiştir. Bir pandeminin sonuçlarını tahmin etmek, politika yapıcılar ve karar vericiler için oldukça önemli ve zor bir görevdir. Bu çalışmanın amacı, çeşitli zaman serisi modelleme yaklaşımlarını kullanarak Türkiye'deki günlük vaka sayılarını tahmin etmektir. Bu kapsamda 11 Mart 2020 ile 24 Aralık 2021 tarihleri arasındaki pozitif vaka sayıları bu çalışmada dikkate alınmıştır. Bu çalışma, kapsadığı gözlem sayısı ile daha az gözlem sayısı ile yapılmış diğer çalışmalardan ayrılmaktadır. Bu çalışmada COVID 19 pandemisi sırasındaki tüm dalgalar daha geniş bir zaman diliminde incelenerek analize dâhil edilmiştir. Ayrıca çalışmamızda makine öğrenmesi algoritmalarının bu algoritmalar ile durum tahmini yapılarak karşılaştırılması yanında, yığınlama yaklaşımı ile kullanılan tüm modellerin tahminleri tek bir model altında birleştirilerek tahmin performansının artırılması hedeflenmiştir. Çalışmamız, incelenen ilgili tüm çalışmaları dikkate alarak, bildiğimiz kadarıyla, bu kadar çok model performansını bir arada değerlendiren ve COVID-19 vaka sayıları üzerinde bir yığınlama modeli oluşturan ilk çalışmadır. Çalışmadan elde edilen bulgular, geliştirilen istifleme modeli ile doğrulanan durum tahminlerinin yüksek doğrulukta yapıldığını ve tüm toplu öğrenme yaklaşımlarının bireysel yöntemlere göre daha iyi sonuçlar verdiğini kanıtlamaktadır.

Anahtar Kelimeler

• COVID-19
• zaman serisi
• tahmin
• makine öğrenimi
• coronavirüs
• genelleştirilmiş yığınlama modelleri

TIME SERIES FORECASTING OF COVID-19 CONFIRMED CASES IN TURKEY WITH STACKING ENSEMBLE MODELS

Öz

Since COVID-19 has spread almost across any country and is a serious threat to mankind, it was declared to be a pandemic by WHO. Forecasting the results of a pandemic is a quite important and difficult task for policy makers and decision makers. The aim of this study is to forecast the daily case numbers in Turkey by using various time series modeling approaches. In this context, positive case numbers between March 11, 2020, and December 24, 2021, were taken into account in this study. This study, with the number of observations it covers, differentiates from other studies which have been conducted with few number of observations. In this study, all the waves during the COVID 19 pandemic were included in the analysis by studying a more extensive time period. Moreover, in our study, along with a comparison of machine learning algorithms by making case forecasting with these algorithms, increasing the forecasting performance was aimed by combining the predictions of all models used with the stacking approach under a single model. By taking all the related studies analyzed into account, our study, as far as we know, is the first one to assess this many model performances together and make a stacking model on COVID-19 case numbers. The findings obtained from the study prove that forecasting of the cases validated via the developed stacking model were made with high accuracy, and all ensemble learning approaches produce better results than individual methods.

Anahtar Kelimeler

• COVID-19
• time-series
• forecasting
• machine learning
• coronavirus
• generalized stacking models

Kaynakça

1. Abdulmajeed, K., Adeleke, M., & Popoola, L. (2020). Online forecasting of COVID-19 cases in Nigeria using limited data. Data in Brief, 30. https://doi.org/10.1016/j.dib.2020.105683
2. Ahmar, A. S., & del Val, E. B. (2020). SutteARIMA: Short-term forecasting method, a case: Covid-19 and stock market in Spain. Science of the Total Environment, 729. https://doi.org/10.1016/j.scitotenv.2020.138883
3. Akay, S., & Akay, H. (2021). Time series model for forecasting the number of COVID-19 cases in Turkey. Turkish Journal of Public Health, 19(2), 140-145. https://doi.org/10.20518/tjph.809201.
4. Al Daoud, E. (2019). Comparison between XGBoost, LightGBM and CatBoost using a home credit dataset. International Journal of Computer and Information Engineering, 13(1), 6-10. https://doi.org/10.5281/zenodo.3607805
5. Ali, M., Khan, D. M., Aamir, M., Khalil, U., & Khan, Z. (2020). Forecasting COVID-19 in Pakistan. PLoS One, 15(11). https://doi.org/10.1371/journal.pone.0242762.
6. Ali, Z., Hussain, I., Faisal, M., Nazir, H. M., Hussain, T., Shad, M. Y., ... & Hussain Gani, S. (2017). Forecasting drought using multilayer perceptron artificial neural network model. Advances in Meteorology, 2017. https://doi.org/10.1155/2017/5681308.
7. Arora, P., Kumar, H., & Panigrahi, B. K. (2020). Prediction and analysis of COVID-19 positive cases using deep learning models: A descriptive case study of India. Chaos, Solitons & Fractals, 139. https://doi.org/10.1016/j.chaos.2020.110017.
8. Biswas, P. K., Islam, M. Z., Debnath, N. C., & Yamage, M. (2014). Modeling and roles of meteorological factors in outbreaks of highly pathogenic avian influenza H5N1. PloS One, 9(6). https://doi.org/10.1371/journal.pone.0098471.

9. Box, G. E., Jenkins, G. M., Reinsel, G. C., & Ljung, G. M. (2015). Time series analysis: Forecasting and control. John Wiley & Sons.
10. Breiman, L. (2001) Random forests. Machine Learning, 45, 5-32. https://doi.org/10.1023/A:1010933404324.
11. Ceylan, Z. (2020). Estimation of COVID-19 prevalence in Italy, Spain, and France. Science of The Total Environment, 729. https://doi.org/10.1016/j.scitotenv.2020.138817.
12. Chandu, V. C. (2020). Time series forecasting of COVID-19 confirmed cases with ARIMA model in the South East Asian countries of India and Thailand: A comparative case study. medRxiv, 2020-05.
13. Chen, K. Y., & Wang, C. H. (2007). Support vector regression with genetic algorithms in forecasting tourism demand. Tourism management, 28(1), 215-226. https://doi.org/10.1016/j.tourman.2005.12.018.
14. Chen, T., & Guestrin, C. (2016). Xgboost: A scalable tree boosting system. In Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining, 785-794, arXiv:1603.02754.
15. Chimmula, V. K. R., & Zhang, L. (2020). Time series forecasting of COVID-19 transmission in Canada using LSTM networks. Chaos, Solitons & Fractals, 135. https://doi.org/10.1016/j.chaos.2020.109864
16. Couronné, R., Probst, P., & Boulesteix, A. L. (2018). Random forest versus logistic regression: A large-scale benchmark experiment. BMC bioinformatics, 19(1), 1-14. https://doi.org/10.1186/s12859-018-2264-5.
17. Dairi, A., Harrou, F., Zeroual, A., Hittawe, M. M., & Sun, Y. (2021). Comparative study of machine learning methods for COVID-19 transmission forecasting. Journal of Biomedical Informatics, 118. https://doi.org/10.1016/j.jbi.2021.103791
18. De Oliveira, L. S., Gruetzmacher, S. B., & Teixeira, J. P. (2021). COVID-19 time series prediction. Procedia Computer Science, 181, 973-980. https://doi.org/10.1016/j.procs.2021.01.254.
19. Dehesh, T., Mardani-Fard, H. A., & Dehesh, P. (2020). Forecasting of covid-19 confirmed cases in different countries with arima models. MedRxiv. https://doi.org/10.1101/2020.03.13.20035345.
20. Ding, G., Li, X., Jiao, F., & Shen, Y. (2020). Brief Analysis of the ARIMA model on the COVID-19 in Italy. medRxiv. https://doi.org/10.1101/2020.04.08.20058636.
21. Džeroski, S., & Ženko, B. (2004). Is combining classifiers with stacking better than selecting the best one?. Machine learning, 54(3), 255-273, https://doi.org/10.1023/B:MACH.0000015881.36452.6e .
22. Earnest, A., Chen, M. I., Ng, D., & Sin, L. Y. (2005). Using autoregressive integrated moving average (ARIMA) models to predict and monitor the number of beds occupied during a SARS outbreak in a tertiary hospital in Singapore. BMC Health Services Research, 5(1), 1-8. https://doi.org/10.1186/1472-6963-5- 36.
23. Fidan, H., & Yuksel, M. E. (2022). A comparative study for determining Covid-19 risk levels by unsupervised machine learning methods. Expert Systems with Applications, 190. https://doi.org/10.1016/j.eswa.2021.116243.
24. Freund, Y., & Schapire, R.E. (1997). A decision-theoretic generalization of on-line learning and anapplication to boosting. J. Comput. Syst. Sci. 55(1), 119–139. https://doi.org/10.1006/jcss.1997.1504.
25. Friedman, J.H. (2001). Greedy function approximation: a gradient boosting machine. Annals of Statistics, 29(5), 1189–1232. https://doi.org/10.1214/aos/1013203451.
26. Gardner, M. W., & Dorling, S. R. (1998). Artificial neural networks (the multilayer perceptron)—a review of applications in the atmospheric sciences. Atmospheric environment, 32(14-15), 2627-2636. https://doi.org/10.1016/S1352-2310(97)00447-0
27. Goyal, R., Chandra, P., & Singh, Y. (2014). Suitability of KNN regression in the development of interaction based software fault prediction models. Ieri Procedia, 6, 15-21. https://doi.org/10.1016/J.IERI.2014.03.004.
28. Guan, P., Huang, D. S., & Zhou, B. S. (2004). Forecasting model for the incidence of hepatitis A based on artificial neural network. World journal of gastroenterology: WJG, 10(24), 3579-3582. https://doi.org/10.3748/wjg.v10.i24.3579.
29. Gunn, S. R. (1998). Support vector machines for classification and regression. ISIS technical report, 14(1), 5-16. https://see.xidian.edu.cn/faculty/chzheng/bishe/indexfiles/new_folder/svm.pdf.
30. Hu, W., Hu, W., & Maybank, S. (2008). Adaboost-based algorithm for network intrusion detection. IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 38(2), 577-583. https://doi.org/10.1109/TSMCB.2007.914695.
31. Huang, G., Wu, L., Ma, X., Zhang, W., Fan, J., Yu, X., ... & Zhou, H. (2019). Evaluation of CatBoost method for prediction of reference evapotranspiration in humid regions. Journal of Hydrology, 574, 1029-1041. https://doi.org/10.1016/j.jhydrol.2019.04.085.
32. Imandoust, S. B., & Bolandraftar, M. (2013). Application of k-nearest neighbor (knn) approach for predicting economic events: Theoretical background. International Journal of Engineering Research and Applications, 3(5), 605-610. https://www.scopus.com/record/display.uri?eid=2-s2.0- 84934906871&origin=inward&txGid=194a430bb8abb0e83d36a946950d48e1&featureToggles=FEA TURE_NEW_DOC_DETAILS_EXPORT:1
33. Jabeur, S. B., Gharib, C., Mefteh-Wali, S., & Arfi, W. B. (2021). CatBoost model and artificial intelligence techniques for corporate failure prediction. Technological Forecasting and Social Change, 166. https://doi.org/10.1016/j.techfore.2021.120658.
34. Kaastra, I., & Boyd, M. (1996). Designing a neural network for forecasting financial and economic time series. Neurocomputing, 10(3), 215-236. https://doi.org/10.1016/0925-2312(95)00039-9.
35. Kane, M. J., Price, N., Scotch, M., & Rabinowitz, P. (2014). Comparison of ARIMA and Random Forest time series models for prediction of avian influenza H5N1 outbreaks. BMC bioinformatics, 15(1), 1-9. https://doi.org/10.1016/j.jbi.2021.103791
36. Karcıoğlu, A. A., Tanışman, S., & Bulut, H. (2021). Time series forecasting of COVID-19 transmission in Turkey using ARIMA model and LSTM Network. Avrupa Bilim ve Teknoloji Dergisi, (32), 288-297. https://doi.org/10.31590/ejosat.1039394. (In Turkish)
37. Katris, C. (2021). A time series-based statistical approach for outbreak spread forecasting: Application of COVID19 in Greece. Expert systems with applications, 166. https://doi.org/10.1016/j.eswa.2020.114077.
38. Khan, M., Mehran, M. T., Haq, Z. U., Ullah, Z., Naqvi, S. R., Ihsan, M., & Abbass, H. (2021). Applications of artificial intelligence in COVID-19 pandemic: A comprehensive review. Expert systems with applications, 185. https://doi.org/10.1016/j.eswa.2021.115695.
39. Koçak, M. (2020). A comparison of time-series models in predicting COVID-19 cases. Türkiye Klinikleri Biyoistatistik, 12(1), 89-96. https://doi.org/10.5336/biostatic.2020-75402.
40. Kumar, N. & Susan, S. (2020). COVID-19 pandemic prediction using time series forecasting models. In: 2020 11th International Conference on Computing, Communication and Networking Technologies (ICCCNT). IEEE, 1-7, https://doi.org/10.1109/ICCCNT49239.2020.9225319
41. Lai, D. (2005). Monitoring the SARS epidemic in China: a time series analysis. J Data Sci, 3(3), 279-93. https://doi.org/10.6339/JDS.2005.03(3).229.
42. Li, X., Wang, L., & Sung, E. (2005). A study of AdaBoost with SVM based weak learners. In Proceedings. 2005 IEEE International Joint Conference on Neural Networks, IEEE, 1, 196-201. https://doi.org/ 10.1109/IJCNN.2005.1555829
43. Liashchynskyi, P., & Liashchynskyi, P. (2019). Grid search, random search, genetic algorithm: A big comparison for NAS. arXiv. https://doi.org/10.48550/arXiv.1912.06059%20Focus%20to%20learn%20more
44. Maleki, M., Mahmoudi, M. R., Wraith, D., & Pho, K. H. (2020). Time series modelling to forecast the confirmed and recovered cases of COVID-19. Travel medicine and infectious disease, 37. https://doi.org/10.1016/j.tmaid.2020.101742.
45. McCluskey, W. J., McCord, M., Davis, P. T., Haran, M., & McIlhatton, D. (2013). Prediction accuracy in mass appraisal: a comparison of modern approaches. Journal of Property Research, 30(4), 239-265. https://doi.org/10.1080/09599916.2013.781204.
46. Naimi, A. I., & Balzer, L. B. (2018). Stacked generalization: an introduction to super learning. European journal of epidemiology, 33(5), 459-464. https://doi.org/10.1007/s10654-018-0390-z.
47. Özen, N. S., Saraç, S., & Koyuncu, M. (2021). Prediction of COVID-19 Cases in the United States of America with Machine Learning Algorithms. Avrupa Bilim ve Teknoloji Dergisi, (22), 134-139. https://doi.org/10.31590/ejosat.855113Abstract. (In Turkish)
48. Pavlyshenko, B. (2018). Using stacking approaches for machine learning models. In 2018 IEEE Second International Conference on Data Stream Mining & Processing (DSMP), IEEE, 255-258. https://doi.org/10.1109/DSMP.2018.8478522.
49. Papastefanopoulos V., Linardatos P., & Kotsiantis S. (2020). COVID-19: A comparison of time series methods to forecast percentage of active cases per population. Applied Sciences, 10(11), 3880. https://doi.org/10.3390/app10113880.
50. Petropoulos, F., Makridakis, S., & Stylianou, N. (2022). COVID-19: Forecasting confirmed cases and deaths with a simple time series model. International Journal of Forecasting, 38, 439-452. https://doi.org/10.1016/j.ijforecast.2020.11.010
51. Pontoh, R. S., Zahroh, S., Hidayat, Y., Aldella, R., Jiwani, N. M., & Firman, S. (2020). Covid-19 modelling in South Korea using a time series approach. Int. J. Adv. Sci. Technol, 29(7), 1620-1632. http://sersc.org/journals/index.php/IJAST/article/view/16246.
52. Prokhorenkova, L., Gusev, G., Vorobev, A., Dorogush, A. V., & Gulin, A. (2018). CatBoost: Unbiased boosting with categorical features. Advances in Neural İnformation Processing Systems, 31. https://doi.org/10.48550/arXiv.1706.09516%20Focus%20to%20learn%20more
53. Purwandari, T., Zahroh, S., Hidayat, Y., Sukonob, S., Mamat, M., & Saputra, J. (2022). Forecasting model of COVID-19 pandemic in Malaysia: An application of time series approach using neural network. Decision Science Letters, 11(1), 35-42. https://doi.org/10.5267/j.dsl.2021.10.001
54. Ribeiro, M. H. D. M., da Silva, R. G., Mariani, V. C., & dos Santos Coelho, L. (2020). Short-term forecasting COVID-19 cumulative confirmed cases: Perspectives for Brazil. Chaos, Solitons & Fractals, 135, 109853. https://doi.org/10.1016/j.chaos.2020.109853
55. Qi, Y. (2012). Random forest for bioinformatics. In Ensemble machine learning. Springer, MA, 307-323, https://doi.org/10.1007/978-1-4419-9326-7_11
56. Schapire, R. E. (2013). Explaining adaboost. In Empirical inference. Springer, Berlin, Heidelberg, 37-52. https://doi.org/10.1007/978-3-642-41136-6_5
57. Seo, D. K., Kim, Y. H., Eo, Y. D., Park, W. Y., & Park, H. C. (2017). Generation of radiometric, phenological normalized image based on random forest regression for change detection. Remote Sensing, 9(11). https://doi.org/10.3390/rs9111163.
58. Sevli, O., & Gülsoy, V. G. B. (2020). Machine learning based case estimation using prophet model with time series data for covid-19 outbreak. Avrupa Bilim ve Teknoloji Dergisi, (19), 827-835, https://doi.org/10.31590/ejosat.766623. (In Turkish)
59. Shahriar, S. A., Kayes, I., Hasan, K., Hasan, M., Islam, R., Awang, N. R., ... & Salam, M. A. (2021). Potential of ARIMA-ANN, ARIMA-SVM, DT and CatBoost for atmospheric PM2. 5 forecasting in Bangladesh. Atmosphere, 12(1), 100. https://doi.org/10.3390/atmos12010100.
60. Shastri, S., Singh, K., Kumar, S., Kour, P., & Mansotra, V. (2020). Time series forecasting of Covid-19 using deep learning models: India-USA comparative case study. Chaos, Solitons & Fractals, 140, https://doi.org/10.1016/j.chaos.2020.110227
61. Singh, S., Parmar, K. S., Kumar, J., & Makkhan, S. J. S. (2020). Development of new hybrid model of discrete wavelet decomposition and autoregressive integrated moving average (ARIMA) models in application to one month forecast the casualties cases of COVID-19. Chaos, Solitons & Fractals, 135, 109866. https://doi.org/10.1016/j.chaos.2020.109866.
62. Smola, A. J., & B. Schölkopf, (1998). On a kernel–based method for pattern recognition, regression, approximation and operator ınversion. Algorithmica 22, 211– 231. Technical Report 1064. https://doi.org/10.1007/PL00013831.
63. Sumi, A., Luo, T., Zhou, D., Yu, B., Kong, D., & Kobayashi, N. (2013). Time-series analysis of hepatitis A, B, C and E infections in a large Chinese city: Application to prediction analysis. Epidemiology & Infection, 141(5), 905-915. https://doi.org/10.1017/S095026881200146X.
64. Talkhi, N., Fatemi, N. A., Ataei, Z., & Nooghabi, M. J. (2021). Modeling and forecasting number of confirmed and death caused COVID-19 in IRAN: A comparison of time series forecasting methods. Biomedical Signal Processing and Control, 66. https://doi.org/10.1016/j.bspc.2021.102494.
65. Tandon, H., Ranjan, P., Chakraborty, T., & Suhag, V. (2020). Coronavirus (COVID-19): ARIMA based timeseries analysis to forecast near future. arXiv preprint arXiv:2004.07859. https://doi.org/10.48550/arXiv.2004.07859.
66. Taud, H., & Mas, J. F. (2018). Multilayer perceptron (MLP). In Geomatic approaches for modeling land change scenarios. Springer, 451-455. https://doi.org/10.1007/978-3-319-60801-3_27.
67. Vapnik, V. (1995). The Nature of Statistical Learning Theory. Springer https://books.google.com.tr/books?hl=tr&lr=&id=sna9BaxVbj8C&oi=fnd&pg=PR7&dq=).+The+Nat ure+of+Statistical+Learning+Theory&ots=oqL9H_jrc6&sig=a5xoMNgbvj_3GrI92a2kcUHDPE&redir_esc=y#v=onepage&q=).%20The%20Nature%20of%20Statistica l%20Learning%20Theory&f=false.
68. Walsh, B. (2017). The world is not ready for the next pandemic. https://time.com/4766624/next-global-security/
69. Wolpert, D. H. (1992). Stacked generalization. Neural networks, 5(2), 241-259. https://doi.org/10.1016/S0893- 6080(05)80023-1
70. Ying, C., Qi-Guang, M., Jia-Chen, L., & Lin, G. (2013). Advance and prospects of AdaBoost algorithm. Acta Automatica Sinica, 39(6), 745-758. https://doi.org/10.1016/S1874-1029(13)60052-X.
71. Zeroual, A., Harrou, F., Dairi, A., & Sun, Y. (2020). Deep learning methods for forecasting COVID-19 time-Series data: A Comparative study. Chaos, Solitons & Fractals, 140, https://doi.org/10.1016/j.chaos.2020.110121.
72. Zhang, Y., & Haghani, A. (2015). A gradient boosting method to improve travel time prediction. Transportation Research Part C: Emerging Technologies, 58, 308–324. https://doi.org/10.1016/j.trc.2015.02.019.
73. Zhao, Y., Chetty, G., & Tran, D. (2019). Deep Learning with XGBoost for Real Estate Appraisal, In 2019 IEEE Symposium Series on Computational Intelligence (SSCI), IEEE, pp. 1396-1401, Xiamen- China, December, https://doi.org/10.1109/SSCI44817.2019.9002790.