Klasifikasi Penyakit Padi Dengan Menggunakan Algoritma Swarm Optimization Untuk Optimasi Metode Backpropagation

Abstract

Rice is the main source of food for the people of Indonesia. In this study, the Backpropagation method in Artificial Neural Networks (ANN) was used to classify rice plant diseases based on observed symptoms. However, the Backpropagation method has weaknesses in convergence speed and accuracy in classifying. To overcome this weakness, the Particle Swarm Optimization (PSO) algorithm is used to optimize the Backpropagation weights. The data used in this study was taken from farmers in the Samarinda area and includes symptoms of rice plant diseases. Various combinations of PSO parameters have been tested to find the optimal configuration. The results showed that the use of PSO for Backpropagation optimization significantly improved the accuracy of rice plant disease classification. The accuracy of the Backpropagation method without using optimization is 79%, while after being optimized using PSO, the accuracy increases to 86%. This research has proven that the use of the PSO algorithm can increase the effectiveness and accuracy of the Backpropagation method in the classification of rice plant diseases.