Article: Prediction of weather using high-performance gradient boosting Journal: International Journal of Global Warming (IJGW) 2023 Vol.31 No.1 pp.30 - 41 Abstract: Our weather prediction technology is imprecise despite its many new uses. Thus, demand exists to adopt a new method that eliminates the system's drawbacks and accurately projects rain. Existing machine learning methods use more RAM, are hard to trim, take a long time to compute, and are hard to use for time series predicting datasets. A high-performance gradient-boosting framework-based decision tree algorithm predicts rain. We used light gradient boosting machine (Light GBM), a leaf-wise method with best-fitting models that eliminates overfitting better than other decision tree algorithms. Predicting continuous goal variables is faster, more efficient, and uses less memory. Rain is Seattle's trademark. This study uses the Seattle dataset of daily weather from 1948 to 2017. The goal is to compute DATE, PRCP, TMAX, TMIN, and RAIN at each break and create a final forecast based on the sampled light BGM that is more accurate than other boosting algorithms. Inderscience Publishers - linking academia, business and industry through research

Title: Prediction of weather using high-performance gradient boosting

Authors: V. Bibin Christopher; R. Isaac Sajan; T.S. Akhila; M. Joselin Kavitha

Addresses: Department of Computing Technologies, SRM Institute of Science and Technology, Kattankulathur Campus, Chengalpattu Dt., Tamilnadu, India ' Department of Electronics and Communication Engineering, Ponjesly College of Engineering, Alamparai, Tamilnadu, India ' Department of Electronics and Communication Engineering, Mar Ephraem College of Engineering and Technology, Lavuvilai, Marthandam, Tamilnadu, India ' Department of Electronics and Communication Engineering, Marthandam College of Engineering and Technology, Marthandam, Tamilnadu, India

Abstract: Our weather prediction technology is imprecise despite its many new uses. Thus, demand exists to adopt a new method that eliminates the system's drawbacks and accurately projects rain. Existing machine learning methods use more RAM, are hard to trim, take a long time to compute, and are hard to use for time series predicting datasets. A high-performance gradient-boosting framework-based decision tree algorithm predicts rain. We used light gradient boosting machine (Light GBM), a leaf-wise method with best-fitting models that eliminates overfitting better than other decision tree algorithms. Predicting continuous goal variables is faster, more efficient, and uses less memory. Rain is Seattle's trademark. This study uses the Seattle dataset of daily weather from 1948 to 2017. The goal is to compute DATE, PRCP, TMAX, TMIN, and RAIN at each break and create a final forecast based on the sampled light BGM that is more accurate than other boosting algorithms.

Keywords: light gradient boosting machine; light GBM; leaf-wise algorithm; precipitation; PRCP; temperature maximum; TMAX; temperature minimum; TMIN; weather forecasting.

DOI: 10.1504/IJGW.2023.133219

International Journal of Global Warming, 2023 Vol.31 No.1, pp.30 - 41

Received: 29 Dec 2022
Accepted: 21 Feb 2023
Published online: 01 Sep 2023 *

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

Title: Prediction of weather using high-performance gradient boosting

Keep up-to-date