Abstract
Temporal knowledge graphs (TKGs) extend conventional knowledge graphs (KGs) by incorporating temporal information to represent dynamic relationships over time. Event evolution mining leverages TKGs to uncover temporal patterns, causality, and sequence of events in complex systems. With applications spanning social media analysis, finance, cybersecurity, and healthcare, TKGs and event evolution mining have emerged as crucial tools for understanding evolving phenomena. This paper presents a comprehensive review of TKGs, their construction, embedding techniques, temporal reasoning models, and event evolution mining methods. We discuss challenges such as data sparsity, reasoning over long-term dependencies, and scalability, alongside current solutions. Future research directions are highlighted to address limitations and enhance the utility of temporal knowledge systems.
Keywords: Temporal knowledge graphs, event evolution, temporal reasoning, dynamic knowledge graphs, temporal embeddings, causal inference, time aware graph modeling.
Full Issue
| View or download the full issue | PDF 41-53 |