Research and Application of PSO-Based BP Neural Network in the Project Estimate of Government Investment

In order to increase the accuracy of the cost estimates in government investment, a method based on the PSO trained neural network to estimate the cost is proposed. First the neural network model of a project cost estimate is created, and then PSO is introduced to optimize the weight and threshold of the neural network, at last the neural network trained is used to estimate cost of the project. The results show that this method not only overcome the defects of traditional neural network such as learning for a long time and shock, but also increase the accuracy of the cost of the project to estimate.