IMPLEMENTASI METODE CONVOLUTIONAL NEURAL NETWORK UNTUK KLASIFIKASI CITRA SEL MALARIA DENGAN ARSITEKTUR INCEPTION-V3

This study focused on classifying malaria cell images. Malaria is a dangerous disease caused by the bite of a female anopheles mosquito. This disease can cause death if the patient is not immediately given appropriate and quick action. The purpose of this study is to provide good accuracy values by implementing the Convolutional Neural Network model using the Inception-V3 architecture as a way to detect malaria automatically with a computer system so that it is more efficient. The Convolutional Neural Network model is a popular method used for classification. This method is a development of Multilayer Perceptron (MLP) which is designed to process two-dimensional data or image data. The Inception-V3 model as a part model of CNN has a more complex architecture and efficient computing and is able to produce smaller model sizes with faster calculation processing times. In this study, training was carried out on 27,558 malaria cell image data using the CNN method with the Inception-V3 architecture by proposing 3 skenario. The proposed scenario 1 model is to use the SGD optimizer to produce a loss value of 0.13 and an accuracy value of 0.95, in scenario 2 an adam optimizer is used which produces a loss value of 0.09 and an accuracy value of 0.96, then scenario 3 implements an RMSprop optimizer resulting in a loss value of 0.08 and an accuracy value of 0.97. Judging from these three scenarios, it can be seen that the results of the study show that the Inception-V3 model using the RMSprop optimizer is able to provide the best accuracy results compared to scenario 1 and scenario 2, which obtained an accuracy of 97% with the loss value so that this study provides better performance than previous studies