Comparative study: Using machine learning techniques about rainfall prediction

Global warming has an impact on people all around the globe, which has a substantial impact on them and is hastening climate change. An accurate forecasting system is required for early detection and enhanced agricultural land management. Rainfall forecasting is a challenging job in reality, and the...

Full description

Saved in:
Bibliographic Details
Main Authors: Hasan R.A., Alomari M.F., Jamaluddin J.B.
Other Authors: 58487876600
Format: Conference Paper
Published: American Institute of Physics Inc. 2024
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.uniten.dspace-34053
record_format dspace
spelling my.uniten.dspace-340532024-10-14T11:17:47Z Comparative study: Using machine learning techniques about rainfall prediction Hasan R.A. Alomari M.F. Jamaluddin J.B. 58487876600 57350402200 58695052500 Global warming has an impact on people all around the globe, which has a substantial impact on them and is hastening climate change. An accurate forecasting system is required for early detection and enhanced agricultural land management. Rainfall forecasting is a challenging job in reality, and the findings must be precise. Rainfall forecasting is a major problem because of the unreliability of existing methods for predicting rainfall. The purpose of this paper to build an accurate model for the daily prediction of the rainfall in Australia using Python In order to find the optimal model based on testing accuracy, four machine learning algorithms are utilised for training and testing (Logistic Regression, Gaussian Naive Bayes, XGboost classifier, and Random Forest). The data was collected at a variety of meteorological stations around Australia using Kaggle. This data collection has 145460 records and 22 attributes. As a result, in order to find the best suitable technique, a comparative study was undertaken after applying four methodologies. Furthermore, a variety of techniques are used, including multiple linear regression and support vector regression, which may provide the most accurate results (up to 78 %). We observed that our model is less accurate when we compared it to this study. With an accuracy of 84.61 %, the XGBoost classifier surpasses rival approaches. There are several misconceptions, according to the matrix. It will be necessary to combine diverse methodologies in the future to increase prediction accuracy. In addition, new datasets are being used, and new areas are being explored. � 2023 Author(s). Final 2024-10-14T03:17:47Z 2024-10-14T03:17:47Z 2023 Conference Paper 10.1063/5.0148472 2-s2.0-85176736708 https://www.scopus.com/inward/record.uri?eid=2-s2.0-85176736708&doi=10.1063%2f5.0148472&partnerID=40&md5=6cc5911327b4d9175dc6341ae57302cb https://irepository.uniten.edu.my/handle/123456789/34053 2787 1 50014 American Institute of Physics Inc. Scopus
institution Universiti Tenaga Nasional
building UNITEN Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Tenaga Nasional
content_source UNITEN Institutional Repository
url_provider http://dspace.uniten.edu.my/
description Global warming has an impact on people all around the globe, which has a substantial impact on them and is hastening climate change. An accurate forecasting system is required for early detection and enhanced agricultural land management. Rainfall forecasting is a challenging job in reality, and the findings must be precise. Rainfall forecasting is a major problem because of the unreliability of existing methods for predicting rainfall. The purpose of this paper to build an accurate model for the daily prediction of the rainfall in Australia using Python In order to find the optimal model based on testing accuracy, four machine learning algorithms are utilised for training and testing (Logistic Regression, Gaussian Naive Bayes, XGboost classifier, and Random Forest). The data was collected at a variety of meteorological stations around Australia using Kaggle. This data collection has 145460 records and 22 attributes. As a result, in order to find the best suitable technique, a comparative study was undertaken after applying four methodologies. Furthermore, a variety of techniques are used, including multiple linear regression and support vector regression, which may provide the most accurate results (up to 78 %). We observed that our model is less accurate when we compared it to this study. With an accuracy of 84.61 %, the XGBoost classifier surpasses rival approaches. There are several misconceptions, according to the matrix. It will be necessary to combine diverse methodologies in the future to increase prediction accuracy. In addition, new datasets are being used, and new areas are being explored. � 2023 Author(s).
author2 58487876600
author_facet 58487876600
Hasan R.A.
Alomari M.F.
Jamaluddin J.B.
format Conference Paper
author Hasan R.A.
Alomari M.F.
Jamaluddin J.B.
spellingShingle Hasan R.A.
Alomari M.F.
Jamaluddin J.B.
Comparative study: Using machine learning techniques about rainfall prediction
author_sort Hasan R.A.
title Comparative study: Using machine learning techniques about rainfall prediction
title_short Comparative study: Using machine learning techniques about rainfall prediction
title_full Comparative study: Using machine learning techniques about rainfall prediction
title_fullStr Comparative study: Using machine learning techniques about rainfall prediction
title_full_unstemmed Comparative study: Using machine learning techniques about rainfall prediction
title_sort comparative study: using machine learning techniques about rainfall prediction
publisher American Institute of Physics Inc.
publishDate 2024
_version_ 1814061164074631168
score 13.211869