Reproduce “EventKG: A Multilingual Event-Centric Temporal Knowledge Graph” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Simon Gottschalk, Elena Demidova
EventKG, developed at the L3S Research Center, establishes a multilingual event-centric temporal knowledge graph by integrating diverse structured and semi-structured sources. This resource identifies over 690,000 events, increases temporal information coverage for 76.21% of events, and introduces metrics for event popularity and relation strength.
Key insights
- A dedicated event-centric approach that systematically integrates diverse structured and semi-structured sources can substantially improve the coverage and completeness of event representations, especially regarding temporal and spatial information.
- Extending a generic event model with enhanced temporal relation types and quantitative popularity/strength metrics provides a more nuanced and application-ready knowledge graph for temporal reasoning and analysis.
- An open-source, extensible generation framework ensures the sustainability and adaptability of a complex multilingual temporal knowledge graph, facilitating future updates, language additions, and data source integrations.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.