Speaker：Jingjing Cao（Wuhan University of Technology）

Time：2022-11-25, 14:30

Location：Tencent Meeting ID：579-922-4321（No Password）

Abstract:

Data Mining is one of the important research directions in the field of e-commerce, especially in the prediction of e-commerce sales. In the actual scenario of e-commerce platform, users, goods and merchants and the data generated among them contain rich information and heterogeneous features, how to consider the impact of various characteristics on sales forecast, and tap the hidden needs of users, are the urgent problems to be solved. This report will introduce four data mining-based e-commerce sales forecasting methods from the multi-dimensional perspectives of customer portrait, customer segmentation, and sentiment analysis for the two tasks of predicting the purchase and repurchase behavior of e-commerce users. These methods are designed to efficiently extract user or product features to improve the accuracy and interpretability of predictive models.