Dynamic Network-Based Two-Stage Time Series Forecasting for Affiliate Marketing

2025-10-13

AI Summary

In affiliate marketing, accurately predicting promoters’ indirect contributions—specifically their propagation scale—remains challenging. To address this, we propose a dynamic-network-driven two-stage time-series forecasting method. First, we introduce a novel propagation scale metric to quantify promoters’ cascading influence. Second, we design a decoupled framework separating network topology from node signals: Stage I employs descendant-neighbor graph convolution to model local propagation dynamics; Stage II leverages hypergraph convolution to capture higher-order promotional relationships. Additionally, we incorporate three auxiliary tasks—self-sales prediction, descendant propagation forecasting, and promoter activation modeling—to enhance robustness and generalization. Evaluated on a large-scale industrial dataset, our method demonstrates significant improvements. After deployment on Alibaba Mom’s platform, it achieved a 9.29% increase in GMV and a 5.89% uplift in sales volume.

Technology Category

• Data Mining & Knowledge Management: Recommender Systems
• Search and Optimization: Algorithm Configuration
• Machine Learning: Learning Preferences or Rankings

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems

Abstract

In recent years, affiliate marketing has emerged as a revenue-sharing strategy where merchants collaborate with promoters to promote their products. It not only increases product exposure but also allows promoters to earn a commission. This paper addresses the pivotal yet under-explored challenge in affiliate marketing: accurately assessing and predicting the contributions of promoters in product promotion. We design a novel metric for evaluating the indirect contributions of the promoter, called propagation scale. Unfortunately, existing time series forecasting techniques fail to deliver accurate predictions due to the propagation scale being influenced by multiple factors and the inherent complexities arising from dynamic scenarios. To address this issue, we decouple the network structure from the node signals and propose a two-stage solution: initially, the basic self-sales and network structure prediction are conducted separately, followed by the synthesis of the propagation scale. Specifically, we design a graph convolution encoding scheme based on descendant neighbors and incorporate hypergraph convolution to efficiently capture complex promotional dynamics. Additionally, three auxiliary tasks are employed: self-sales prediction for base estimations, descendant prediction to synthesize propagation scale, and promoter activation prediction to mitigate high volatility issues. Extensive offline experiments on large-scale industrial datasets validate the superiority of our method. We further deploy our model on Alimama platform with over $100,000$ promoters, achieving a $9.29%$ improvement in GMV and a $5.89%$ increase in sales volume.

Problem

• Accurately predicting promoter contributions in affiliate marketing
• Decoupling network structure from node signals for forecasting
• Addressing dynamic promotional complexities with multi-task learning

Innovation

• Two-stage forecasting decouples network and node signals
• Graph convolution encoding captures complex promotional dynamics
• Auxiliary tasks mitigate volatility and improve prediction accuracy