PSO-Optimized BP Neural Network for Classification
A program implementing Particle Swarm Optimization (PSO) to enhance Backpropagation Neural Networks for classification tasks. The implementation follows a two-phase approach: first using PSO to optimize initial weights and thresholds, then training the BP network with momentum and adaptive learning rate algorithms. The attached materials include dataset and modular functions for data extraction, target generation, baseline BP implementation, PSO optimization, and integrated PSO-BP training.