Beyond Isolated Clients: Integrating Graph-Based Embeddings into Event Sequence Models
This work addresses the need for better fraud prevention and recommendations in digital platforms by enhancing event sequence models with graph embeddings, though it is incremental as it builds on existing SSL methods.
The paper tackles the problem of predicting user attributes from timestamped interactions by integrating graph-based structural information into self-supervised learning models, resulting in improved accuracy up to 2.3% AUC across financial and e-commerce datasets.
Large-scale digital platforms generate billions of timestamped user-item interactions (events) that are crucial for predicting user attributes in, e.g., fraud prevention and recommendations. While self-supervised learning (SSL) effectively models the temporal order of events, it typically overlooks the global structure of the user-item interaction graph. To bridge this gap, we propose three model-agnostic strategies for integrating this structural information into contrastive SSL: enriching event embeddings, aligning client representations with graph embeddings, and adding a structural pretext task. Experiments on four financial and e-commerce datasets demonstrate that our approach consistently improves the accuracy (up to a 2.3% AUC) and reveals that graph density is a key factor in selecting the optimal integration strategy.