用图神经网络预测多门店销量，效果优于传统方法。
Leveraging Spatiotemporal Graph Neural Networks for Multi-Store Sales Forecasting
- 构建动态图模型捕捉门店间关联，通过残差路径预测对数差分销量。
- 在45家沃尔玛门店数据上，误差最低，尤其在高波动门店表现更稳。
- 无需地理信息就能发现功能相似门店群，适合连锁零售场景。
本文评估了时空图神经网络（STGNN）在多门店零售销售预测中的有效性，并与ARIMA、LSTM和XGBoost基线模型进行比较。基于45家沃尔玛门店的周度销售数据，构建了一个通过学习自适应图来建模门店间依赖关系的预测框架。该模型对数差分后的销量进行预测，并通过残差路径重构最终值，实现稳定训练与更好泛化能力。实验表明，STGNN在归一化总绝对误差、P90 MAPE以及各门店MAPE方差三项指标上均优于所有基线。对学习到的邻接矩阵分析显示，即使无地理元数据，也能识别出有意义的功能门店聚类及高影响力节点。结果表明，关系结构显著提升互联零售环境下的预测质量，确立了STGNN在多门店需求预测中的稳健性。
原文摘要 · Abstract (English)
This work evaluates the effectiveness of spatiotemporal Graph Neural Networks (GNNs) for multi-store retail sales forecasting and compares their performance against ARIMA, LSTM, and XGBoost baselines. Using weekly sales data from 45 Walmart stores, we construct a relational forecasting framework that models inter-store dependencies through a learned adaptive graph. The proposed STGNN predicts log-differenced sales and reconstructs final values through a residual path, enabling stable training and improved generalisation. Experiments show that STGNN achieves the lowest overall forecasting error, outperforming all baselines in Normalised Total Absolute Error, P90 MAPE, and variance of MAPE across stores. Analysis of the learned adjacency matrix reveals meaningful functional store clusters and high-influence nodes that emerge without geographic metadata. These results demonstrate that relational structure significantly improves forecast quality in interconnected retail environments and establishes STGNNs as a robust modelling choice for multi-store demand prediction.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。