hybrid levenberg marquardt and back propagation neural network for house price prediction in taiwan

Price prediction is an influential tool in each market to enhance economic performance. in this regard, regression methods are often used. however, with the expansion of artificial intelligence, neural networks, machine learning, and deep learning, these methods can also be used for prediction. on the other hand, pricing in the housing market is always challenging, and forecasting the house price has been one of the concerns of economic activists in this field. accordingly, in this research, a hybrid levenberg marquardt (lm) and back propagation (bp) neural network has been developed to forecast housing prices in taiwan. this artificial intelligence method can provide a suitable forecast for housing price trends in the future by using the information in the form of input and output. this method uses inputs such as inflation rate, bank interest rate, minimum wage, and gross domestic product (gdp). moreover, the housing price index is considered as the output of the model. in order to implement the proposed method, data from taiwan from 1998 to 2022 was used. in this regard, a percentage of this data is used as training data, and the rest is used as test data in the artificial neural network. the results show that the rmse of the proposed method is less than classic lm and bp methods. finally, the proposed neural network will achieve the final housing price in taiwan from 2023 through 2027. the results show that housing prices will trend upward in this country in the next five years.