Title: Efficient data pruning using optimal KNN for weather forecasting in cloud computing

Authors: T. Benil; P. Krishna Kumar; R. Bharathi

Addresses: School of Computer Science and Engineering, Vellore Institute of Technology, Chennai Campus, India ' Department of Computer Science and Engineering, Amrita College of Engineering and Technology, Erachakulam, Kanyakumari District, India ' Department of Electronics and Communication Engineering, University College of Engineering, Nagercoil, India

Abstract: Established researchers may use cloud-deployed software, information sharing, collaborative effort devices, and cloud-based computer infrastructure to handle data and recreate models. This study examines cloud nomenclature in the meteorological local region to aid numerical weather forecast. Numerical weather prediction forecasts weather. Numerous machine-learning methods are suggested for numerical weather prediction equation evaluation (NWP). Despite these inaccuracies, the methodologies so far provide decent results. It is hard to manage weather history, but it is best to do so. Thus, we propose a novel k-nearest neighbour classifier to prune missing values. Despite these inaccuracies, the methodologies so far provide decent results. The new KNN, optimum k-nearest neighbour, uses an enhanced way to choose out the precise data required for prediction. Seattle rainfall datasets assess performance. We found that the optimum KNN method improves classification accuracy while reducing time.

Keywords: K-nearest neighbour; numerical weather prediction; NWP; cloud computing; weather forecasting.

DOI: 10.1504/IJGW.2023.130984

International Journal of Global Warming, 2023 Vol.30 No.2, pp.137 - 151

Received: 29 Aug 2022
Accepted: 20 Sep 2022
Published online: 17 May 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article