WEATHER FORECASTING SYSTEM USING GAUSSIAN NAÏVE BAYES

Abstract

Forecasting is the process of estimation in unknown situations from the historical data. Weather forecasting is one of the applications to predict the atmosphere state for a future time and a given location as regards temperature, cloudiness, dryness, wind, rain and etc. Weather forecasting system is implemented as the responsive web application depends on OpenWeather API. The point of the proposed system is to classify the climate class using Gaussian Naïve Bayes based on weather information acquired from this API. The architecture of the proposed system consists of API from Open Weather, MySQL database for data storage of historical weather dataset, and PHP programming component. In this system, 63648 weather records are contained in the weather database and then 52608 records are used as training data and 11040 are testing. The system has achieved nearly 89% accuracy using training data. For testing data, the system has achieved nearly 79% accuracy. According to the analysis outcomes, the proposed weather system can classify for climate status with the highest accuracy.