Wikipedia Event Reporter
Extracting Event-Related Information From the Wikipedia Edit History

Wikipedia is a multilingual, web-based, free-content encyclopedia project based on an openly editable model. It is the most up-to-date encyclopedia available. Consequently, as new events take place all over the world, Wikipedia users will update the articles corresponding to the entities involved in these events, or influenced by them, causing an avalanche of edits on several articles, as more information regarding the event becomes available. Because Wikipedia has well defined standards for the quality of its articles, and it strives to provide a point of view as neutral as possible, in the course of the event, through a mutual agreement process the different points of view converge to view accepted by the community of contributors.
As an encyclopedia, Wikipedia covers much of what is of importance for a general reader. As it does not have periodical releases, and is a live encyclopedia, Wikipedia has to be constantly updated when events influences the subject. The large community of users is always mobilized and takes care of keeping the accuracy and actuality of the information. To make sense of all the collaborative editing involved, all revisions of an article are kept, in an edit history. We use this as our news source. A burst of edits indicates a period with an increased level of attention over one article. Investigating the bursts, we find the source of the increased activity, and if this is an event, we analyze the updates that took place inside the burst, in order to provide a comprehensive summary of the event.
In order to detect and extract event-specific information from Wikipedia updates, we use the methods presented in our scientific paper. The data set used in the paper is now available.
 



is an advanced application for detecting event-related information from the stream of Wikipedia edits. More information about all the functionalities of the Wikipedia Edit Reporter are available in the Man page. Details about how the magic happens and the tools used are available in the Behind the Scenes page.




Particular showcase:
European Politicians Events Detector - 12
European Politicians Events Detector - 10
European Politicians Events Detector - 8

Papers:
Extracting Event-Related Information from Article Updates in Wikipedia
Temporal Summarization of Event-Related Updates in Wikipedia (demonstration at WWW'2013)

Dataset:
Labeled Updates



Thank you for your interest and support!

The working team:
Mihai Georgescu
Dang Duc Pham
Sergej Zerr
Nattiya Kanhabua
Stefan Siersdorfer
Contact the authors