Research on real estate pricing methods based on data mining and machine learning

This paper aims to study the actual utility of real estate pricing models based on data mining and machine learning. In order to achieve this goal, this paper introduces appropriate trend estimation methods, adjusts pricing models and processes, and realizes trend estimation that changes over time to make the resulting pricing model have advantages such as dynamics, accuracy, and flexibility over the original model. Moreover, this paper proposes a real estate pricing model based on the quadratic exponential smoothing time-varying trend estimation. In addition, this paper uses the quadratic exponential smoothing method to calculate the index trend in the random process of the house price index, and performs segmentation processing on the pricing model according to the index cycle, so as to obtain the real estate pricing method under the time-varying trend. Finally, this paper adjusts the volatility parameters and the mean recovery rate to time-varying piecewise functions, uses the quadratic variation to calculate the volatility parameters and the martingale valuation method to calculate the mean recovery rate parameters, and establishes the real estate pricing model of the time-varying O–U process. Case studies show that the model constructed in this paper has good performance and has certain practical effects.