Tingkat Ketelitian Pengenalan Pola Data pada Algoritma Pelatihan Perbaikan Metode Batch Mode dalam Jaringan Syaraf Tiruan

Abstract

Backpropagation method in neural network using some training algorithms in problem solving. These algorithms need to be tested to get the most accuracy in identifying patterns of data. In this study conducted testing of 6 training algorithms that included in the improved of batch mode algorithms, i.e. traingda, traingdx, trainrp, trainbfg, trainoss, and trainlm. Based on the results of the statistical tests using analysis of variance (ANOVA) with a confidence level of 95% of the obtained results that trainlm algorithm is the most accurate with an average error of 0,0063. Thus these results can be used as a basis for the development of research and applications in the field of neural networks specifically for researchers or educators for development of science and technology