Forschungspraktikum Open Knowledge Bases

Sommersemester 2016

Are you interested in Free Knowledge? Do you like data management? This summer term you will have the opportunity to learn about Open Knowledge Bases and work with Wikidata, one of the latest projects of Wikimedia, the organisation that manages Wikipedia.

Wikidata

Wikidata is a free and collaborative knowledge base created and maintained by volunteers. Wikidata represents human knowledge, but in contrast to Wikipedia, it contains structured data. Wikidata describes world entities (e.g. locations, persons, events, books, etc.) via statements that contain claims (e.g. “Koblenz has sister city Maastricht”) and references to external information sources that support the claims.

As an example, you can see the Wikidata item for Koblenz here:

https://www.wikidata.org/wiki/Q3104

Wikidata can be used by any system and it is becoming the database for Wikipedia. Wikidata’s centralized and multilingual data management can make Wikipedia articles improve their consistency and information coverage across different languages. For example, if the Infoboxes of the articles for Koblenz in the different Wikipedias (e.g. English, German, Spanish) point to the Wikidata item for Koblenz, when one updates the population of Koblenz in Wikidata, that number will be simultaneously updated in all Wikipedia articles and will save a lot of work to editors who would not need to edit each of the different language Wikipedia articles. See this presentation by Krötzsch for further motivational examples of Wikidata.  

One of the particularities of Wikidata is that it allows to represent knowledge from different points of view, as Pintscher pointed out : a Wikidata item may have statements with multiple values. Wikidata can specify that an item (e.g. Berlin) has a particular value (e.g. 3,000,000 inhabitants) for a property (e.g. population), according to a particular source of information (e.g. the statistics office of Berlin); while a different source specifies that the item has a different value for the same property (e.g. 3,500,000 inhabitants). Moreover, multiple values of statements may be ranked.

Goal of the research lab

The aim of this research lab is to work on methods and applications that deal with multi-provenance knowledge in Wikidata. We will work with a use case scenario around news and events. On the one hand, we will work on a tool for gathering multi-view facts about events and encourage their curation in Wikidata. On the other hand, we will define and analyze methods to semi-automatically rank the collected multi-view facts according to Wikidata guidelines.

Team Work

We will organize the Projektpraktikum together with the Forschungspraktikum. Bachelor students will focus on software engineering tasks, while master students will work on research-oriented tasks. Students will work in teams and they are encouraged to use agile methodologies (e.g. Scrum) and collaborative software development technology (e.g. GitHub).

Introductory Meeting

We will have an initial meeting to describe the goals and the methodology of the course in detail.

Agenda

• Open Knowledge Bases

• Goals and Administrative Issues of the Course

• Technical Introduction to Wikidata

When?

11.04.2016 at 10:00 CET

Where?

Building C- Room 208

Registration

IMPORTANT: Registration is now closed.

Students interested in attending the Projektpraktikum and Forschungspraktikum are invited to come to the introductory meeting.

In order to make the organization of the meeting easier, if you are interested and you plan to attend please register before 08.04.2016 via this form: https://docs.google.com/forms/d/1a_xz-v7Mqt26st37nhDUHvQevvxmF7jiJs5SRbPANMg/viewform

Contact

Cristina Sarasua <csarasua@uni-koblenz.de>

Meetings

Thu 14.04 at 12:00 E-428

Materials

Wikidata Links

Wikidata Wiki: https://www.wikidata.org/wiki/Wikidata:Main_Page

Wikidata Documentation: https://www.wikidata.org/wiki/Help:Contents

Wikidata Data Access: https://www.wikidata.org/wiki/Wikidata:Data_access

Wikidata API: https://www.wikidata.org/w/api.php

Wikibase Data Model: https://www.mediawiki.org/wiki/Wikibase/DataModel

Wikidata Classes and Properties: http://tools.wmflabs.org/wikidata-exports/miga/?classes#

Wikidata - All Properties: https://www.wikidata.org/wiki/Wikidata:List_of_properties/all

Wikidata RDF Data Model: https://www.mediawiki.org/wiki/Wikibase/Indexing/RDF_Dump_Format#Ontology

Bibliographic References

Krötzsch, M., Vrandečić, D., & Völkel, M. (2006). Semantic Mediawiki. In: The Semantic Web-ISWC 2006 (pp. 935-942). Springer Berlin Heidelberg. URL: http://c.unik.no/images/6/6d/SemanticMW.pdf  

Denny Vrandečić and Markus Krötzsch. 2014. Wikidata: a free collaborative knowledgebase.Commun. ACM 57, 10 (September 2014), 78-85. DOI="http://dx.doi.org/10.1145/2629489" URL: http://korrekt.org/papers/Wikidata-CACM-2014.pdf

Erxleben, F., Günther, M., Krötzsch, M., Mendez, J., & Vrandečić, D. (2014). Introducing Wikidata to the linked data web. In The Semantic Web–ISWC 2014 (pp. 50-65). Springer International Publishing. URL: http://korrekt.org/papers/Wikidata-RDF-export-2014.pdf

Müller-Birn, C., Karran, B., Lehmann, J., & Luczak-Rösch, M. (2015, August). Peer-production system or collaborative ontology engineering effort: What is Wikidata?. In Proceedings of the 11th International Symposium on Open Collaboration (p. 20). ACM. URL: http://www.opensym.org/os2015/proceedings-files/p501-mueller-birn.pdf

Lehrende

Cristina Sarasua

• Alumnus