A Survey of Link Prediction in Temporal Networks

2025-02-28

AI Summary

Temporal Link Prediction (TLP) lacks a unified methodological framework, hindering systematic comparison and advancement. Method: This work proposes the first explicit taxonomy that decouples *representation learning* from *reasoning mechanisms*, systematically organizing techniques—including dynamic graph neural networks, temporal encoding, sequential modeling, and statistical inference—across dimensions of dynamic structural modeling, transductive/inductive settings, and scalability challenges. Contributions: (1) Introduces a novel two-dimensional classification scheme—Representation × Reasoning—that clarifies the functional boundaries and assumptions of existing TLP methods; (2) Identifies underexplored yet promising cross-paradigm combinations (e.g., inductive reasoning over learned temporal embeddings); and (3) Establishes a foundational framework for interpretable modeling and scalable TLP on large-scale temporal networks. This taxonomy fills a critical gap in TLP’s theoretical infrastructure, enabling principled design of highly scalable and interpretable models.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Planning, Routing, and Scheduling: Temporal Planning
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Temporal networks have gained significant prominence in the past decade for modelling dynamic interactions within complex systems. A key challenge in this domain is Temporal Link Prediction (TLP), which aims to forecast future connections by analysing historical network structures across various applications including social network analysis. While existing surveys have addressed specific aspects of TLP, they typically lack a comprehensive framework that distinguishes between representation and inference methods. This survey bridges this gap by introducing a novel taxonomy that explicitly examines representation and inference from existing methods, providing a novel classification of approaches for TLP. We analyse how different representation techniques capture temporal and structural dynamics, examining their compatibility with various inference methods for both transductive and inductive prediction tasks. Our taxonomy not only clarifies the methodological landscape but also reveals promising unexplored combinations of existing techniques. This taxonomy provides a systematic foundation for emerging challenges in TLP, including model explainability and scalable architectures for complex temporal networks.

Problem

• Temporal Link Prediction (TLP) forecasts future connections in dynamic networks.
• Existing surveys lack a comprehensive framework for representation and inference methods.
• The paper introduces a taxonomy to classify and analyze TLP approaches systematically.

Innovation

• Novel taxonomy for Temporal Link Prediction
• Classification of representation and inference methods
• Exploration of unexplored technique combinations