High Performance Implementation of Neural Networks Learning Using Swarm Optimization Algorithms for EEG Classification Based on Brain Wave Data

EEG analysis aims to help scientists better understand the brain, physicians diagnose and treatment choices of brain-computer interface. Artificial neural networks are among most effective learning algorithms perform computing tasks similar biological neurons in human brain. In some problems, network model's performance might significantly degrade overfit due irrelevant features that negatively influence model performance. Swarm optimization robust techniques can be implemented find optimal solutions such problems. this paper, Grey Wolf Optimizer (GWO) Particle Optimization (PSO) applied for feature selection training a Feed-forward Neural Network (FFNN). The FFNN terms test accuracy, precision, recall, F1_score is investigated. Furthermore, research has other five machine purpose comparison. Experimental results prove outperforms all via GWO.