Modeling and Simulation of Real Estate Valuation Based on Genetic Neural Network Model

Abstract

China’s real estate market is developing rapidly, but the house price is abnormal. The nonlinear relationship between housing characteristics and real estate value is difficult to calculate, resulting in the difficulty of house price prediction. Based on this, the relationship model between characteristic variables and house prices is constructed by using the machine learning method. At the same time, genetic algorithm is used to screen the specific values. The experimental results show that the optimized model converges in 56 iterations; in the application test, the research model found that in 90% of the test samples, the error between the predicted value and the actual measured value shall not exceed 10%. Experiments show that the genetic algorithm is effective in optimizing the BP house price valuation model and improves the calculation efficiency and valuation accuracy of the valuation model.