Pengembangan Model Pembelajaran Mesin Untuk Mengidentifikasi Nominal Uang Kertas Rupiah Menggunakan Algoritma CNN Dengan Library Keras

Abstract

This study develops application to identify the value of rupiah banknotes. The rupiah banknotes used are the 2009 and 2016 issuances. The purpose of this study is to produce a machine learning model that can be used as a basis for identifying the value of rupiah banknotes and creating a GUI to be able to use its functions easily. The method used in this study is the Convolutional Neural Network. The results obtained are models that have a final value of training loss = 7%, training accuracy = 98%, validation loss = 33%, and validation accuracy = 93%. The finished model has also been tested and can identify new image data.