| Autor: |
Nramban Kannan, Senthil Kumar, Kolla, Bhanu Prakash, Sengan, Sudhakar, Muthusamy, Rajendiran, Manikandan, Raja, Patel, Kanubhai K., Dadheech, Pankaj |
| Zdroj: |
SN Computer Science; January 2024, Vol. 5 Issue: 1 |
| Abstrakt: |
Sustainable development is crucial for a prosperous future, but epidemic diseases like Coronavirus Disease 2019 (COVID-19) pose real and complex challenges. The global pandemic, declared by the WHO on March 11, 2020, has led to significant loss of life and challenges in public health, economy, food systems, and social life. This research article aims to develop a Machine Learning (ML) model to predict a patient’s COVID-19 diagnosis, as early diagnosis is fundamental for controlling the spread of COVID-19. An investigation of the literature and a research study is conducted to find suitable mathematical methods and measure how they impact prediction models. This paper analyses COVID-19 pandemic deaths, confirmed cases, and recovered individuals using Time Series Analysis (TSA) to study the disease's impacts and understand the TAS. A forecasting model can predict future COVID cases by analyzing trends in time-series and connecting global changes with government restrictions. Since higher predictive accuracy is the limitation of ensemble learning algorithms, better ML approaches are proposed here. Autocorrelation plots clearly showed the results executed for the considered objectives. The hybrid ARIMA algorithm proposed in this work proved adequate results. |
| Databáze: |
Supplemental Index |
| Externí odkaz: |
|
Full text from SpringerLink