Research on Product Sales Forecasting under Paid Traffic Scenarios on E-commerce Platforms
The rapid development of e-commerce and increasing competition make accurate sales forecasting crucial for companies to reduce inventory costs and optimize marketing strategies.Currently,paid traffic has become a favored option for merchants on e-commerce platforms.However,there is limited research on the impact of paid traffic on sales,and there is a lack of detailed studies on the effects of paid traffic from different channels on sales.Additionally,there is no solution for assessing the quality of traffic from these channels.To refine the study of the impact of paid traffic from different channels on sales and improve the accuracy of sales forecasts,we propose the DCA-OLSTM model.Using real data from the Taobao platform,experiments show that incorporating paid traffic features can improve the accuracy of sales forecasts,and that paid traffic from different channels has varying degrees of impact on sales predictions.Merchants can choose paid traffic channels with higher weight values.Finally,comparative experiments demonstrate that the proposed model outperforms other models in terms of accuracy.
paid traffice-commercesales forecastingLSTMtraffic by channel
万方数据
维普
