Editorial
Rainfall Prediction using XGB Model with the Australian Dataset
@ARTICLE{10.4108/ew.5386, author={Surendra Reddy Vinta and Radhika Peeriga}, title={Rainfall Prediction using XGB Model with the Australian Dataset}, journal={EAI Endorsed Transactions on Energy Web}, volume={11}, number={1}, publisher={EAI}, journal_a={EW}, year={2024}, month={3}, keywords={Rainfall prediction, Machine Learning, Randomforest, Logistic regression, Gaussian Naive Bayes, K-nearest neighbors, Support Vector classifier, XGBoost, Accuracy, Cross-validation, Performance metrics, Australian rainfall dataset, Data preprocessing, Feature selection, Missing value imputation, Practical applications}, doi={10.4108/ew.5386} }- Surendra Reddy Vinta
Radhika Peeriga
Year: 2024
Rainfall Prediction using XGB Model with the Australian Dataset
EW
EAI
DOI: 10.4108/ew.5386
Abstract
Rainfall prediction is a critical field of study with several practical uses, including agriculture, water management, and disaster preparedness. In this work, we examine the performance of several machine learning models in forecasting rainfall using a dataset of Australian rainfall observations from Kaggle. Six models are compared: random forest (RF), logistic regression (LogReg), Gaussian Naive Bayes (GNB), k-nearest neighbours (kNN), support vector classifier (SVC), and XGBoost (XGB). Missing value imputation and feature selection were used to preprocess the dataset. To analyse the models, we employed cross-validation and performance indicators such as accuracy, precision, recall, and F1-score. According to our findings, the RF and XGB models fared the best, with accuracy ratings of 87% and 85%, respectively. With accuracy ratings below 70%, the GNB and SVC models performed the poorest. Our findings imply that machine learning algorithms can be useful tools for predicting rainfall, but careful model selection and evaluation are required for reliable results.
Copyright © 2024 S. R. Vinta et al., licensed to EAI. This is an open access article distributed under the terms of the CC BY-NC-SA 4.0, which permits copying, redistributing, remixing, transformation, and building upon the material in any medium so long as the original work is properly cited.
