Wikimedia Research Showcase - December 18, 2019

YouTube stream:

Conversations in IRC on freenode, #wikimedia-research channel.

Making Knowledge Bases More Complete

By Fabian Suchanek, Télécom Paris, Institut Polytechnique de Paris

A Knowledge Base (KB) is a computer-readable collection of facts about the world (examples are Wikidata, DBpedia, and YAGO). The problem is that these KBs are often missing entities or facts. In this talk, I present some new methods to combat this incompleteness. I will also quickly talk about some other research projects we are currently pursuing, including a new version of YAGO. Publications

The Dynamics of Peer-Produced Political Information During the 2016 U.S. Presidential Campaign

By Brian Keegan, Ph.D., Assistant Professor, Department of Information Science, University of Colorado Boulder

Wikipedia plays a crucial role for online information seeking and its editors have a remarkable capacity to rapidly revise its content in response to current events. How did the production and consumption of political information on Wikipedia mirror the dynamics of the 2016 U.S. Presidential campaign? Drawing on systems justification theory and methods for measuring the enthusiasm gap among voters, this paper quantitatively analyzes the candidates’ biographical and related articles and their editors. Information production and consumption patterns match major events over the course of the campaign, but Trump-related articles show consistently higher levels of engagement than Clinton-related articles. Analysis of the editors’ participation and backgrounds show analogous shifts in the composition and durability of the collaborations around each candidate. The implications for using Wikipedia to monitor political engagement are discussed. Paper