Penerapan JST Untuk Prakiraan Cuaca Di Wilayah Kota Blitar Menggunakan Metode Algoritma Hopfield

Abstract

Weather forecasting is the application of science and technology to estimate a state of the atmosphere in the future and will occur in a certain area. In this study, weather forecasting in the Blitar City area uses the Hopfield Algorithm Artificial Neural Network which can be used to identify patterns of numbers and letters using classification techniques. The weather criteria used in this method are wind speed, temperature, humidity and air pressure. The collected morning, afternoon and evening data are classified according to the expected target output, i.e. rainy, sunny and cloudy. From the collection of 21 data that was carried out, it resulted in a match with the target weather of 76% accuracy and an error rate of 24%.