An E-Commerce Coupon Target Population Positioning Model Based on Random Forest and eXtreme Gradient Boosting

Abstract

At present, the commonly used e-commerce coupon target population location method is based on Logistic, of which the positioning accuracy is not high in the case of serious data loss. In this paper, we propose a complex classification model based on Random Forest (RF)and eXtreme Gradient Boosting (XGBoost), and test the reliability of it through experiments. Our experimental results show that the model has good performance on the online Alibaba O2O Coupon Usage Forecast competition dataset.