Neuro Rainfall Forecast with Data Mining by Real-Coded Genetical Preprocessing

Abstract

In this paper, rainfall is predicted by using a Neural Network(NN) and a Genetic Algorithm(GA). GA selects data needed to predict the rainfall. NN learns and predicts it using attributes selected by GA. The real-coded GA is used to decide data priority, and data really needed for the rainfall forecast are selected based on the priority. In order to show the effectiveness of the proposed rainfall prediction system, computer simulations are performed for real weather data. Finally, the effectiveness of this system is shown with data analysis.