Time series forecasting of new cases and new deaths rate for COVID-19 using deep learning methods

被引:68
|
作者
Ayoobi, Nooshin [1 ]
Sharifrazi, Danial [2 ]
Alizadehsani, Roohallah [3 ]
Shoeibi, Afshin [4 ,5 ]
Gorriz, Juan M. [6 ]
Moosaei, Hossein [7 ]
Khosravi, Abbas [3 ]
Nahavandi, Saeid [3 ]
Chofreh, Abdoulmohammad Gholamzadeh [8 ]
Goni, Feybi Ariani [9 ]
Klemes, Jiri Jaromir [8 ]
Mosavi, Amir [10 ,11 ]
机构
[1] Savitribai Phule Pune Univ, Dept Math, Pune 411007, Maharashtra, India
[2] Islamic Azad Univ, Sch Tech & Engn, Dept Comp Engn, Shiraz, Iran
[3] Deakin Univ, Inst Intelligent Syst Res & Innovat IISRI, Waurn Ponds, Vic 3217, Australia
[4] Ferdowsi Univ Mashhad, Comp Engn Dept, Mashhad, Razavi Khorasan, Iran
[5] KN Toosi Univ Technol, Biomed Data Acquisit Lab, Fac Elect & Comp Engn, Tehran, Iran
[6] Univ Granada, Dept Signal Theory Networking & Commun, Granada, Spain
[7] Univ Bojnord, Fac Sci, Dept Math, Bojnurd, Iran
[8] Brno Univ Technol VUT Brno, Fac Mech Engn, NETME Ctr, Sustainable Proc Integrat Lab SPIL, Tech 2896-2, Brno 61669, Czech Republic
[9] Brno Univ Technol VUT Brno, Fac Business & Management, Dept Management, Kolejni 2906-4, Brno 61200, Czech Republic
[10] Obuda Univ, John von Neumann Fac Informat, H-1034 Budapest, Hungary
[11] Norwegian Univ Life Sci, Sch Business & Econ, N-1430 As, Norway
关键词
Long Short Term Memory (LSTM); Convolutional Long Short Term Memory (Conv-LSTM); Gated Recurrent Unit (GRU); Bidirectional New Cases of COVID-19; New Deaths of COVID-19; COVID-19; Prediction; Deep learning; Machine learning; MODEL;
D O I
10.1016/j.rinp.2021.104495
中图分类号
T [工业技术];
学科分类号
08 ;
摘要
The first known case of Coronavirus disease 2019 (COVID-19) was identified in December 2019. It has spread worldwide, leading to an ongoing pandemic, imposed restrictions and costs to many countries. Predicting the number of new cases and deaths during this period can be a useful step in predicting the costs and facilities required in the future. The purpose of this study is to predict new cases and deaths rate one, three and seven-day ahead during the next 100 days. The motivation for predicting every n days (instead of just every day) is the investigation of the possibility of computational cost reduction and still achieving reasonable performance. Such a scenario may be encountered in real-time forecasting of time series. Six different deep learning methods are examined on the data adopted from the WHO website. Three methods are LSTM, Convolutional LSTM, and GRU. The bidirectional extension is then considered for each method to forecast the rate of new cases and new deaths in Australia and Iran countries. This study is novel as it carries out a comprehensive evaluation of the aforementioned three deep learning methods and their bidirectional extensions to perform prediction on COVID-19 new cases and new death rate time series. To the best of our knowledge, this is the first time that Bi-GRU and Bi-Conv-LSTM models are used for prediction on COVID-19 new cases and new deaths time series. The evaluation of the methods is presented in the form of graphs and Friedman statistical test. The results show that the bidirectional models have lower errors than other models. A several error evaluation metrics are presented to compare all models, and finally, the superiority of bidirectional methods is determined. This research could be useful for organisations working against COVID-19 and determining their long-term plans.
引用
收藏
页数:15
相关论文
共 50 条
  • [1] Forecasting COVID-19 new cases using deep learning methods
    Xu, Lu
    Magar, Rishikesh
    Farimani, Amir Barati
    COMPUTERS IN BIOLOGY AND MEDICINE, 2022, 144
  • [2] COVID-19: Forecasting confirmed cases and deaths with a simple time series model
    Petropoulos, Fotios
    Makridakis, Spyros
    Stylianou, Neophytos
    INTERNATIONAL JOURNAL OF FORECASTING, 2022, 38 (02) : 439 - 452
  • [3] Deep learning methods for forecasting COVID-19 time-Series data: A Comparative study
    Zeroual, Abdelhafid
    Harrou, Fouzi
    Dairi, Abdelkader
    Sun, Ying
    CHAOS SOLITONS & FRACTALS, 2020, 140
  • [4] Modelling and forecasting of new cases, deaths and recover cases of COVID-19 by using Vector Autoregressive model in Pakistan
    Khan, Firdos
    Saeed, Alia
    Ali, Shaukat
    CHAOS SOLITONS & FRACTALS, 2020, 140
  • [5] Forecasting COVID-19 cases using dynamic time warping and incremental machine learning methods
    Miralles-Pechuan, Luis
    Kumar, Ankit
    Suarez-Cetrulo, Andres L.
    EXPERT SYSTEMS, 2023, 40 (06)
  • [6] Evaluation of Bayesian Deep Learning Methods for Quantifying Uncertainties in Forecasting Deaths from COVID-19
    André Vinícius Gonçalves
    Gustavo Medeiros de Araújo
    Eugênio Monteiro da Silva Júnior
    SN Computer Science, 6 (3)
  • [7] A novel approach based on combining deep learning models with statistical methods for COVID-19 time series forecasting
    Abbasimehr, Hossein
    Paki, Reza
    Bahrini, Aram
    NEURAL COMPUTING & APPLICATIONS, 2022, 34 (04): : 3135 - 3149
  • [8] A novel approach based on combining deep learning models with statistical methods for COVID-19 time series forecasting
    Hossein Abbasimehr
    Reza Paki
    Aram Bahrini
    Neural Computing and Applications, 2022, 34 : 3135 - 3149
  • [9] Forecasting COVID-19 Cases in Morocco: A Deep Learning Approach
    Hankar, Mustapha
    Birjali, Marouane
    Beni-Hssane, Abderrahim
    NETWORKING, INTELLIGENT SYSTEMS AND SECURITY, 2022, 237 : 845 - 857
  • [10] Approach to COVID-19 time series data using deep learning and spectral analysis methods
    Oshinubi, Kayode
    Amakor, Augustina
    Peter, Olumuyiwa James
    Rachdi, Mustapha
    Demongeot, Jacques
    AIMS BIOENGINEERING, 2022, 9 (01): : 1 - 21