This month’s Research showcase will be held tomorrow, Thursday, Dec. 18th at 3PM PST (2300
UTC). As usual, the event will be recorded and publicly streamed on YouTube (link
<https://www.youtube.com/watch?v=xPO8XhmeUAU>) We’ll hold a discussion and take
questions from the Wikimedia Research IRC channel (#wikimedia-research
<http://webchat.freenode.net/?channels=wikimedia-research> on freenode).
Looking forward to seeing you there.
Dario
——
This month:
Mobile Madness: The Changing Face of Wikimedia Readers
By Oliver Keyes <https://www.mediawiki.org/wiki/User:Ironholds>
A dive into the data we have around readership that investigates the rising popularity of
the mobile web, countries and projects that are racing ahead of the pack, and what changes
in user behaviour we can expect to see as mobile grows.
Global Disease Monitoring and Forecasting with Wikipedia
By Reid Priedhorsky <http://www.lanl.gov/expertise/profiles/view/reid-priedhorsky>
(Los Alamos National Laboratory)
Infectious disease is a leading threat to public health, economic stability, and other key
social structures. Efforts to mitigate these impacts depend on accurate and timely
monitoring to measure the risk and progress of disease. Traditional, biologically-focused
monitoring techniques are accurate but costly and slow; in response, new techniques based
on social internet data, such as social media and search queries, are emerging. These
efforts are promising, but important challenges in the areas of scientific peer review,
breadth of diseases and countries, and forecasting hamper their operational usefulness. We
examine a freely available, open data source for this use: access logs from the online
encyclopedia Wikipedia. Using linear models, language as a proxy for location, and a
systematic yet simple article selection procedure, we tested 14 location-disease
combinations and demonstrate that these data feasibly support an approach that overcomes
these challenges. Specifically, our proof-of-concept yields models with r² up to 0.92,
forecasting value up to the 28 days tested, and several pairs of models similar enough to
suggest that transferring models from one location to another without re-training is
feasible. Based on these preliminary results, we close with a research agenda designed to
overcome these challenges and produce a disease monitoring and forecasting system that is
significantly more effective, robust, and globally comprehensive than the current state of
the art.