Large Scalable Cross-Domain Graph Neural Networks for Personalized Notification at LinkedIn

2025-06-15

AI Summary

To address challenges in professional social platforms—including difficulty in fusing cross-domain heterogeneous signals, weak temporal dynamics modeling, and conflicting multi-objective optimization—this paper proposes a large-scale cross-domain heterogeneous graph neural network framework for notification recommendation. We introduce a novel paradigm for constructing unified cross-domain graphs, modeling users, content, and behavioral signals as a temporally enhanced heterogeneous graph. The framework integrates dynamic graph encoding with multi-task joint learning to enable real-time inference over billion-scale nodes. Online deployment demonstrates significant improvements over single-domain baselines: a 0.10% increase in weekly active users and a 0.62% lift in click-through rate, validating enhanced multi-objective synergy. This work delivers a scalable, interpretable, and low-latency graph learning solution tailored for industrial-grade personalized notification systems.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Application Domains: Social Networks
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: Social recommender systems and personalization for the social Web
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Notification recommendation systems are critical to driving user engagement on professional platforms like LinkedIn. Designing such systems involves integrating heterogeneous signals across domains, capturing temporal dynamics, and optimizing for multiple, often competing, objectives. Graph Neural Networks (GNNs) provide a powerful framework for modeling complex interactions in such environments. In this paper, we present a cross-domain GNN-based system deployed at LinkedIn that unifies user, content, and activity signals into a single, large-scale graph. By training on this cross-domain structure, our model significantly outperforms single-domain baselines on key tasks, including click-through rate (CTR) prediction and professional engagement. We introduce architectural innovations including temporal modeling and multi-task learning, which further enhance performance. Deployed in LinkedIn's notification system, our approach led to a 0.10% lift in weekly active users and a 0.62% improvement in CTR. We detail our graph construction process, model design, training pipeline, and both offline and online evaluations. Our work demonstrates the scalability and effectiveness of cross-domain GNNs in real-world, high-impact applications.

Problem

• Integrating heterogeneous signals across domains for notifications
• Capturing temporal dynamics in user engagement prediction
• Optimizing multiple competing objectives in recommendation systems

Innovation

• Cross-domain GNN integrating user, content, activity
• Temporal modeling and multi-task learning enhancements
• Large-scale graph construction for unified signal processing