A Hyperparameter Optimization for Galaxy Classification

In this study, the morphological galaxy classification process was carried out with a hybrid approach. Since Galaxy may contain detailed information about universe’s formation, it remains current research topic. Researchers divided more than 100 billion galaxies into ten different classes. It is not always possible to understand which class types belong. However, Artificial Intelligence (AI) can be used for successful classification. There are studies on automatic of small number As classes increases, success methods decreases. Based literature, using Convolutional Neural Network (CNN) better. Three meta-heuristic algorithms obtain optimum architecture CNN. These Grey Wolf Optimizer (GWO), Particle Swarm Optimization (PSO) and Bee Colony (ABC) algorithms. A CNN nine hidden layers two full connected used. The neurons in fully layers, learning coefficient batch size values were optimized. accuracy my model 85%. best results obtained GWO. Manual optimization difficult. help GWO algorithm.