Pendugaan Tinggi Pasang Surut Laut Harian Menggunakan Jaringan Syaraf Tiruan Metode Backpropagation

Abstract

Artificial Neural Network Model with methods Backrpopagation authors try to be implemented in the estimation or prediction of high tidal waters in Bengkulu. Neural Network architecture that is built consisting of three layers namely input layer, hidden layer and output layer. Estimation tidal itself always relies on historical data of the place concerned. A person can not make a prediction of ocean tides in a particular region in the absence of historical data in the region in a time sequence. Time series data is used as a basis of the estimation of data so as to recognize the tidal patterns that occur which in turn can be used as a reference to estimate the number of ups and downs that will occur. In this study, preprocessing or initial process carried out by the method of Backpropagation, where there are several steps that lead levels that this initial process can be carried out with the best. Changing the data type in the desired interval by Artificial Neural Networks will determine the next processing step. The sequence of steps taken in this prepocessing is selecting the data, data cleaning, transformation of data