1. Home
  2. Vol 2, No 1 (2023)
  3. Olukunle Oyebode

The Prediction Rate of COVID-19 using Random Forest Approach

Ebenezer Olukunle Oyebode(1), Olufemi Olayanju Awodoye(2), Funmilola Wumi Ipeayeda(3),


(1) Department Computer Science, Faculty of Natural Sciences, Ajayi Crowther University, Oyo, Nigeria
(2) Department of Computer Engineering, Faculty of Engineering, Ajayi Crowther University, Oyo, Nigeria
(3) Department Computer Science, Faculty of Natural Sciences, Ajayi Crowther University, Oyo, Nigeria
Corresponding Author

Abstract


COVID-19 is a pandemic disease that claimed a lot of human lives and caused economic setbacks among other problems. Its effects became more negative and prolonged partly because of poor prediction rates that could give priviledge information for preparation against the disease. A dataset containing several information inform of records of suspected number of cases, deaths, location etc of 338 records were gathered from ECDPC. Random forest tree model was setup in Phython using ANACONDA. The random forest model setup was trained to predict the number of cases and likely number of deaths resulting from such cases within 1-10 days and in weeks. It was observed that the model accuracy in the prediction for the number of cases in days was 99.2% while that of the number of cases in weeks is 99.94%.

Keywords


Pandemic, model, random forest, dataset

Full Text: PDF

Article Metrics

Abstract View : 162 times
PDF Download : 21 times

DOI: 10.56534/acjpas.v2i1.77

DOI (PDF): https://doi.org/10.56534/acjpas.v2i1.77.g18

Refbacks

  • There are currently no refbacks.