Design of Real Estate Enterprise Value Evaluation Model Based on BP Neural Network Algorithm

With the development of urbanization, China’s real estate market has developed rapidly, and a large number of transactions such as consumption and investment have appeared. Among them, residential real estate is a hot spot in the real estate market, accounting for the largest proportion of development investment. Because the real estate industry is a capital-intensive enterprise, there are some problems, such as favoring large-scale development, only focusing on profit growth, excessive equity financing, etc., while the traditional value evaluation method ignores the role of equity capital cost and cannot truly reflect the operating performance and value creation ability of real estate enterprises. However, in the evaluation of real estate listed companies, the applicability of methods, the selection of specific methods and the specific application of models need further study. Secondly, based on the shortcomings of traditional methods, this paper introduces the basic idea of neural network into real estate evaluation, and analyzes the feasibility and advantages of BP neural network (BPNN) in real estate evaluation. Then, based on the theoretical methods in the previous two chapters, the influencing factors of real estate prices were comprehensively analyzed, and an indicator system of influencing factors of residential real estate prices was constructed. The indicators were quantified, standardized, and trended. The importance of the indicators was ranked using random forest theory, reducing the number of indicators, and optimizing the indicator system.