Loading…

Study of short term rain forecasting using machine learning based approach

Weather forecasting has been still dependent on statistical and numerical analysis in most part of the world. Though statistical and numerical analysis provides better results, it highly depends on stable historical relationships with the predict and predicting value of the predict and at a future t...

Full description

Saved in:
Bibliographic Details
Published in:Wireless networks 2021-11, Vol.27 (8), p.5429-5434
Main Authors: Balamurugan, M. S., Manojkumar, R.
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Weather forecasting has been still dependent on statistical and numerical analysis in most part of the world. Though statistical and numerical analysis provides better results, it highly depends on stable historical relationships with the predict and predicting value of the predict and at a future time. On the other hand, machine learning explores new algorithmic approaches in prediction which is based on data-driven prediction. Climatic changes for a location are dependent on variable factors like temperature, precipitation, atmospheric pressure, humidity, wind speed and combination of other such factors which are variable in nature. Since climatic changes are location-based statistical and numerical approaches result in failure at times and needs an alternate method like machine learning based study of understanding about the weather forecast. In this study it has been observed that percentage in departure of rainfall has been ranging from 46 to 91% for the month of June 2019 as per Indian Meteorological Department (IMD) by using the traditional forecasting methods, but whereas based on the following study implemented using machine learning it has been observed that forecast was able to achieve much better rainfall prediction comparative to statistical methods.
ISSN:1022-0038
1572-8196
DOI:10.1007/s11276-019-02168-3