Transaction price prediction of second-hand houses in Wuhan based on GA-BP model

Abstract

The significance of the used housing market to the stability of people's lives and social development is increasing, and the prediction of its prices is becoming a key concern for society. In this paper, BP neural network model is used to predict the price of second-hand properties. In view of the disadvantages of slow convergence and the tendency to obtain local optimal solutions, the BP neural network model's input layer is optimized by a genetic algorithm to speed up the convergence speed and accuracy of the BP neural network model. The experimental results show that the improved GA-BP neural network has high prediction accuracy, with RMSE and MAE of 398.72 and 170.18, respectively. The difference between the actual and predicted values is small, which can provide a more stable tool for predicting house prices in the second-hand property market and broaden a new channel for the practical application of GA-BP neural network.