PERFORMANCE OPTIMIZATION OF NEURAL NETWORK USING GA INCORPORATED PSO

摘要

Present invention provides feed forward neural network (FFNN) optimized by genetic crossover and mutation process (GA) incorporated by particle swarm optimization (PSO) which is applied for optimizing the hidden layer neuron for the neural network. The main pursuit is to reduce the size of the network that can be rehearsed expedite than the simple back propagation algorithm while keeping the same or superior performance. The process is optimized not only number of hidden layers but also the number of neurons in each hidden layer and processing it to deal with a few connections in network structure so as to enhance the speed and efficiency of the neural network. Following invention is described in detail with the help of Figure 1 of sheet 1 showing the flow chart of implemented work.