Subway Station Investment Prediction Based on IGWO-SVR

To predict the investment of subway stations quickly and accurately in the feasibility study stage and provide decision support for investors, this paper proposed a method to optimize the parameters of support vector regression machine based on improved grey wolf optimizer. Firstly, by collecting and sorting out cases of subway stations and establishing a database as the original prediction sample, bootstrapping was adopted to expand the sample to solve the problem of low accuracy of small sample prediction. Secondly, the initial population generation, convergence factor and location update of the grey wolf optimizer were improved to avoid the problem of grey wolf optimizer being stuck in local optima. Subsequently, the improved grey wolf optimizer was used to optimize the parameters of support vector regression machine to establish a subway station investment prediction model. Finally, a station example was used to verify the validity of the prediction model. The results show that the average relative error of the prediction model of the support vector regression machine optimized by the improved grey wolf algorithm is 4.33%, with the goodness of fit of 0.944 0 on the test set. The prediction model based on improved grey wolf algorithm with optimized parameters is better than the unoptimized, particle swarm optimized and grey wolf algorithm optimized support vector regression prediction models. The relative error of the case of Honggaolu station is 4.87%, proving the effectiveness of the proposed prediction method. © 2024 Science Press. All rights reserved.