Mass Appraisal of Real Estate Prices Using Improved BP Neural Network with Policy Evaluation

This paper proposes a new policy index neural network (NN), which uses the back propagation (BP) NN model optimized by genetic algorithm (GA) and particle swarm algorithm and the policy modeling consistency (PMC) index to quantify the relevant real estate policies. This study also trains and validates the big data of primary housing transactions of 33 properties in the main urban area of Weihai City, Shandong Province and evaluates the model performance using mean squared error, mean absolute percentage error, and R
²
indexes. Subsequently, it compares the model performance under the GA and particle swarm algorithm optimization, respectively. The results show that the BPNN model with policies as independent variables is more accurate in predicting house prices in Weihai City.