A Multilingual Event-Centric Temporal Knowledge Graph

The EventKG is a multilingual¹ resource incorporating event-centric information extracted from several large-scale knowledge graphs such as Wikidata, DBpedia and YAGO, as well as less structured sources such as the Wikipedia Current Events Portal and Wikipedia event lists in 15 languages. The EventKG is an extensible event-centric resource modeled in RDF. It relies on Open Data and best practices to make event data spread across different sources available through a common representation and reusable for a variety of novel algorithms and real-world applications.

¹ currently supported languages: English, French, German, Italian, Russian, Portuguese, Spanish, Dutch, Polish, Norwegian, Romanian, Croatian, Slovene, Bulgarian and Danish

/ Dataset VoID description

EventKG Applications

EventKG+BT

Interactive spatio-temporal biographies.

Example biography timeline of Angelina Jolie

EventKG+TL

Cross-lingual event timelines.

Example event timeline of the Brexit

EventKG Datasets

EventKG+Click

1000 semantic event-centric queries and their verbalisations in three languages.

In what event hosted in the United States in 2004 was Lose My Breath recorded?
- The 2004 NBA Playoffs
Which award organized by the Netherlands Film Festival in 1981 was received by Hannah Hoekstra?
- The Golden Calf award

Example questions in EventKG (in English; also available: German and Portuguese)

Event-QA

Language-specific event-centric user interaction traces.

English	German	Russian
Southeast Asian Games	2016 Berlin Truck Attack	2009 Russian Premier League
2017 Southeast Asian Games	German student movement	1993 Russian Top League
2014 United States Elections	2006 Austrian legislativ election	2012-13 Russian Premier League

Events with highest language-specific relevance per language

EventKG Statistics

EventKG provides information for over 1.3 million events and over 4.5 million temporal relations. More than a half of the events (56.25%) originate from the existing knowledge graphs; the others are extracted from semi-structured sources.

Comparison of event representations that originate from different resources in EventKG (Version V3.0 - example queries for EventKG, YAGO and English Wikipedia event lists):

	EventKG	Wikidata	YAGO	DBpedia (en)	DBpedia (fr)	DBpedia (bg)	Wikipedia event lists (en)	Wikipedia event lists (bg)
Events	1,348,561	715,286	262,286	302,274	101,213	7,874	208,727	11,560
known location	496,537	480,616	63,051	13,784	7,253	1,016	0	0
known time	1,151,734	498,930	89,276	31,845	23,161	0	208,727	11,560

Example: EventKG contains 302,274 extracted from the English DBpedia. For 13,784 of these events, EventKG has also extracted at least one location from the English DBpedia.

Cite as

@inproceedings{gottschalk2019eventkg,
   title={{EventKG - the Hub of Event Knowledge on the Web - and Biographical Timeline Generation}},
   author={Gottschalk, Simon and Demidova, Elena},
   year={2019},
   publisher={IOS Press},
   volume={10},
   number={6},
   pages={1039--1070},
   journal={Semantic Web Journal (SWJ)}
}

Contact

Simon Gottschalk (gottschalk@L3S.de)

Acknowledgements

EventKG was built at the L3S Research Center and facilitated by the CLEOPATRA Project. The CLEOPATRA Project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement no. 812997.