A Prognostic Approach For Precipitation Forecast Using Naive Bayes Algorithm

Authors

  • Razeef Mohd , Abeena Mohi-u-din Azad , Anwaar Ahmad Wani , Idrees ahmad Bhat

Abstract

Prognosis of precipitation is essential and challenging job for the researchers. Predicting
accurate precipitation is vital for water resource management in agriculture field and disaster
management and its allied sectors. Rainfall prognosis is a type of weather forecasting which
involves taking down the various aspects of conditions. In the recent past, it has been shown that
data mining techniques along with machine learning algorithms have better performance and
prediction than conventional statistical methods. In this work, we have used Naive Bayes technique
for the rainfall prediction. The weather information of Srinagar, India, is collected from
http:///www.wundergrounds.com website. Five (5) most significant weather attributes for
precipitation prognosis are selected from nine (9) attributes. Analysis and testing the prediction by
Naive Bayes Algorithm is performed by comparing the rainfall prediction results of actual data item
for a particular day using Naive Bayes Algorithm with the observed data item of the collected
weather data set. The results of the Naive Bayes Algorithm are found to be quite accurate and
acceptable

Published

2020-11-01

Issue

Section

Articles