Second-hand Housing Price Evaluation Using A LightGBM Model

Since the 21st century,the second-hand housing market has developed rapidly in China.The second-hand housing market share has surpassed the new housing market in the first tier cities,and the second-hand housing transaction is becoming more and more extensi ve.In the process of second-hand house transaction,it is difficult to determine the transaction price of second-hand house due to different demands of both parties and different expectations for the price of second-hand house,which makes the evaluation of second-hand house price highly valued.Reasonable and effective evaluation of second-hand house price is conducive to shortening the transaction cycle of second-hand house,improving the resource allocation of second-hand house market,and ensuring the residential demand of res idents.At present,in the process of second-hand house transaction in China,the traditional method of price evaluation based on the manual method of professional appraisers is often used.Such method is difficult to collect a large number of reference data,unable to conduct price evaluation in batches,and the results given by different professional appraisers are different,and the economic cost and time cost are very high.In this context,in order to improve the efficiency and accuracy of the price evaluation of second-hand housing,this paper applies a LightGBM algorithm to establish the price evaluation model of second-hand housing,and makes an empirical analysis on the transaction data of second-hand housing in Beijing to demonstrate the feasibility of the price evaluation model of second-hand housing based on LightGBM.Compared with other second-hand housing price evalu ation models,the evaluation model established in this paper is more accurate and more efficient.The innovation of this paper reflects the following two aspects.On the one hand,the LightGBM model is applied to the second-hand housing price evaluation in order to improve the efficiency and accuracy of the second-hand housing price evaluation.Traditional second-hand housing price evaluation method is affected by the experience of the evaluator and work efficiency.To train a qualified evaluator requires the long-term training.The introduction of LightGB M model in second-hand housing price evaluation can effectively improve the efficiency and accuracy of the evaluation.On the other hand,a relatively complete secondary housing price evaluation index system is constructed.In particular,some location and neighborhood related indicators are refined,and related indicators are used for quantitative processing.More influencing factors are considered in the index system,and more comprehensive data are collected by using Internet technology,so as to make the second-hand housing price evaluation more real and objective.Based on the LightGBM model,the second-hand housing price evaluation method enriches the second-hand housing price evaluation index system,satisfies the actual use needs of users through empirical analysis,and improves the efficiency and accuracy of the second-hand housing price evaluation to some certain extent.