Skip to content

ISSN: 2672-4787

Log in Register
SJARD logo Semi-Arid Journal of Academic Research and Development
Research Article

Forecasting the Rate of Malaria Spread in Nigeria Using Feed forward Neural Network Trained with Firefly Algorithm

  • Muhammaad Idris1 ✉
  • Aminu Musa2
  • Alhasssan Idris3
  • Abdulrazaq Isah Abubakar4
  • Muazzam Ibrahim5
  • Jibrin Gambo6
  1. 1Department of Computer Science, Binyaminu Usman Polytechnic, Jigawa State Nigeria
  2. 2Department of Computer Science, Binyaminu Usman Polytechnic, Jigawa State Nigeria
  3. 3Department of Computer Science, Kano University of Science and Technology
  4. 4Department of Computer Science, Binyaminu Usman Polytechnic, Jigawa State Nigeria
  5. 5Department of Computer Science, Binyaminu Usman Polytechnic, Jigawa State Nigeria
  6. 6Department of Forestry, Binyaminu Usman Polytechnic, Jigawa State Nigeria

Abstract

Malaria is one of the leading causes of death from infectious diseases in the world. It is a preventable but treatable disease that is caused by parasites called plasmodium and transmitted by Anopheles mosquito bites. Due to the serious health implications of malaria infection, there is a need to develop forecasting techniques that will serve as early warning signals with timely case detection in regions of dynamic malaria transmission so that preventive and control measures can effectively be implemented. This study is an attempt to model and forecast the incidence of malaria infection in Nigeria using an artificial neural network technique. The study uses annual data for malaria infection cases for the period from 1984 to 2014 in Nigeria. In the experiments, we evaluate the strength of an artificial neural network technique trained by a non-supervised learning algorithm (the Firefly algorithm) and benchmark it against a feed-forward neural network trained by a supervised learning algorithm (the Back-propagation Algorithm). The results obtained were compared with results obtained from Back Propagation (BPROP) trained NNs. FA-based NN performed very well in forecasting the future number of people affected by malaria. The study recommends that the government of Nigeria at all levels, international agencies, and policymakers should embark on strong preventive, curative, and control measures in order to reduce the menace of the future.

Keywords

How to cite

Idris, M., Musa, A., Idris, A., Abubakar, A. I., Ibrahim, M., & Gambo, J. (2022). Forecasting the Rate of Malaria Spread in Nigeria Using Feed forward Neural Network Trained with Firefly Algorithm. Semi-Arid Journal of Academic Research and Development, 5(2), 35–43.

APA 7th BibTeX RIS (EndNote, Mendeley, Zotero)

Related articles