Shengjie Ke

Shengjie Ke

05 December 2023

E-commerce, Logistics demand forecasting, Deep learning.

Logistics demand forecasting is an important business in the e-commerce field, involving various types of data with complex features; deep learning, as a powerful machine learning method, can automatically extract high-level feature representations from data, improving prediction performance. This article introduces three basic deep learning network structures suitable for e-commerce logistics demand forecasting, namely the spatial feature extraction network structure based on convolutional neural network (CNN), the temporal feature extraction network structure based on recurrent neural network (RNN), and the multi-dimensional feature fusion network structure based on attention mechanism. This article also proposes methods to evaluate the prediction performance of these network structures, including data sets, evaluation indicators, benchmark methods and evaluation processes.

© 2023, the Authors. Published by ISCCAC

This is an open access article distributed under the CC BY-NC license