Rent Prediction Based On Stacking Regression Model And Baidu MAP API

Rent forecasting has important implications for both renters and rental management.In this paper,by collecting the data of rent and related influencing factors,after data preprocessing,a variety of regression models for rent prediction are established,and on this basis,an integrated regression model is built to predict the rent.Firstly,the rental housing data of 16 municipal districts in Shanghai was crawled through Lianjia website,and the specific latitude and longitude of the community and the macro factors affecting the rent were obtained through the Baidu Map API address resolution service and location retrieval service.Data preprocessing such as missing value filling,outlier processing,and feature generation were performed on the original data.Then common single regression models,such as multiple linear regression model,Ridge regression model,LASSO regression model,multilayer perceptron model,Gradient Boosting regression model,XGBoost regression model,and random forest regression model for rent prediction were established and the experimental results were compared.Finally,in order to further improve the prediction accuracy,a Stacking ensemble regression model was built as the primary predictor model with single models as the first-layer learners and the linear regression model as the secondary learner.The experimental results show that,compared with the single prediction model and the weighted combination prediction model,the Stacking ensemble model proposed in this paper has the best performance in terms of the root mean square error RMSE,goodness of fit R~2,and average relative error MRE.It has excellent forecasting effect and high forecasting accuracy.The results of the study provide new ideas for rent forecasting,which can provide reference for renters and management departments.