Brief Introduction

E-commerce systems exhibit high dynamism, strong intervention, and complex feedback structures. Traditional correlation-based prediction models struggle to support long-term robust decision-making and are susceptible to distribution drift and selection bias. This theme focuses on the integrated innovation of process mining, causal prediction, and multi-objective optimization in e-commerce systems, promoting the transition from correlation analysis to causality-driven process optimization and collaborative decision-making. By integrating causal graph learning, reinforcement learning, and process mining methods, an interpretable, transferable, and intervenable intelligent decision-making framework is constructed. Simultaneously, a multi-objective optimization mechanism is introduced to systematically balance multiple objectives such as revenue enhancement, user experience, platform fairness, and resource allocation efficiency. This theme primarily concentrates on core scenarios including recommendation systems, advertising deployment, dynamic pricing, user lifecycle management, and operational process optimization. It aims to provide theoretical support and methodological innovation for long-term value creation and robust system operation in complex e-commerce ecosystems.

Topics

1. Causal graph learning and e-commerce behavior structure discovery
2. Application of stable learning and out-of-distribution generalization in e-commerce recommender systems
3. Dynamic pricing and inventory collaborative optimization driven by causal reinforcement learning
4. Counterfactual recommendation and causal intervention decision-making for optimizing the conversion path of live streaming e-commerce
5. Graph neural network modeling and multi-objective cost optimization for cross-border e-commerce supply chain collaboration
6. Causal discovery and robust optimization of e-commerce business processes based on process mining
7. E-commerce multi-objective robust decision optimization driven by LLM under causal fusion
8. Not limited to the above themes: any research exploring the intelligent transformation of e-commerce decision-making