Research On Shenzhen Housing Rent Based On Machine Learning Model

With the rapid development of urbanization in China,the prices of commercial housing keep growing,especially in first tier cities,such as Beijing,Shanghai,Guangzhou and Shenzhen,which keep rising all the way.In order to solve the housing problem,more and more people choose to live in rental house.However,the current housing market in China is dominated by sales,and the leasing market has not yet developed in coordination,and there are still many problems.The problems are as follows:the imbalance between supply and demand in the housing rental market,asymmetric information between landlords and tenants,and the chaos in the rental market have been hindering the development of the rental market.In order to solve the above problems,controlling the price trend of the housing leasing market is the key,so it is particularly important to make reasonable pricing and forecasting of rents.Based on the current background of big data,this paper proposes the use of machine learning models to analyze and predict housing rents,and hope to provide a method with better prediction results for reference in the rental market.This paper uses python to crawl the renting house data of Shenzhen in December 2019,and makes statistical analysis of data.We visualize the cleaned data.From the descriptive statistical analysis of the data,we can intuitively see the relationship between the features,and get the factors which affect the rent.Besides,we give some analyses of the result.And then,we transform the data and select the features,12 features are finally used for models.In this paper we uses support vector regression,random forest,and XGBoost models to train data.In order to optimize the model and improve the accuracy of prediction,we use grid search method to optimize the important parameters of the model.Besides,the average absolute percentage error and the coefficient of determination are used as evaluation criterion of models.By comparing the evaluation criterion results of the selected three machine learning models,it was found that XGBoost has the smallest average absolute percentage error and the largest determination coefficient value after the adjustment.It can be seen that XGBoost can better adapt to the unbalanced data set,and can more accurately predict when the real rent is high or low.It shows that XGBoost model has stronger generalization performance and wider application range,so it is more suitable for rent prediction in rental market.