WikiWarMonitor ICTeCollective
Harnessing ICT - enabled collective social behaviour
This webpage is dedicated to a project called WikiWarMonitor, as a part of the European Comission FP7 FET-Open supported project called ICTeCollective.
Here we release datasets, software, and some results of our "Wikipedia editorial war" detection and ranking algorithm.
Problem: Our general aim in this project is to locate, rank, characterise, predict and find ways to resolve Wikipedia edit warrings. To this end, we define a controversy measure M, which quantifies size and intensity of the edit warring taking place at a certain time and on a certain article. Using human judgment, we calibrate M such that the articles with an aggregated M larger/smaller than a threshold, could be considered as controversial/peacful articles. For a short description, please see this paper, and for more details please take a look at our publications listed below.
Media Coverage:
MSNBC: Wikipedia is editorial warzone, says study
BBC Spanish: Las guerras de Wikipedia
Decoded Science: Wikipedia Wars: Implications for Building Consensus
University of Oxford: Mathematical model 'describes' how online conflicts are resolved
Phys.org: Conflicts in Wikipedia now modelled by statistical physicists
Procedure: Our code starts by parsing the Wikipedia dumps available by Wikimedia Foundation to produce what we call "light dumps" which only contains the timestamp , user name, a revision ID which is uniquely assigned to versions with the similar text hash-code. This makes us able to perform many different analysis and measurements on articles history needless of handling the Huge Wikipedia dumps. In the next step we use a separate code to calculate the controversy measure M based on the light dumps produced in the previous step.
Datasets:
Wikipedia dumps
Light dumps
Software:
Code for extracting the light dumps from Wikipedia dumps
Code for parsing the light dumps to calculate M

Yet an improved python implementation to parse the light dumps and calculate M (by H. Sepehri Rad).
Results:
List of top-100 most controversial articles: ar, cs, de, en, es, pt, fa, fr, he, hu, ro, ja, zh.
Complete list of M: ar, cs, de, en, es, pt, fa, fr, he, hu, ro, ja, zh.
List of articles with "never-ending-wars" in English Wikipedia.
Publications:
Our papers on Wikipedia editorial wars:
1) The Most Controversial Topics in Wikipedia: A Multilingual and Geographical Analysis
T. Yasseri, A. SPoerri, M. Graham, and J. Kertész,
In: Fichman P., Hara N., editors. Global Wikipedia: International and Cross-Cultural Issues in Online Collaboration. Scarecrow Press (2014), Forthcoming.
2) Dynamics of conflicts in Wikipedia
T. Yasseri, R. Sumi, A. Rung, A. Kornai, and J. Kertész,
PLoS ONE 7(6): e38869 (2012).
3) Edit wars in Wikipedia
R. Sumi, T. Yasseri, A. Rung, A. Kornai, and J. Kertész,
IEEE Third International Confernece on Social Computing (SocialCom) 9-11 Oct. 724-727, Boston, MA, USA (2011).
4) Opinions, Conflicts and Consensus: Modeling Social Dynamics in a Collaborative Environment
J. Török, G. Iñiguez, T. Yasseri, M. San Miguel, K. Kaski and J. Kertész,
Phys. Rev. Lett. 110(8), 088701 (2013).
Our other papers on Wikipedia:
1) A practical approach to language complexity: a Wikipedia case study,
T. Yasseri A. Kornai and J. Kertész,
arXiv:1204.2765v1 (2012).
2) Circadian patterns of Wikipedia editorial activity: A demographic analysis,
T. Yasseri, R. Sumi and J. Kertész,
PLoS ONE 7(1): e30091 (2012).
3) Early Prediction of Movie Box Office Success based on Wikipedia Activity Big Data
M. Mestyán, T. Yasseri, and J. Kertész,
Submitted (2012).
4) Value production in a collaborative environment
T. Yasseri and J. Kertész,
To appear in J. Stat. Phys. (2013).
Figures:
Michael Jackson
Time evolution of the controversy measure of the article about
Michael Jackson.
A: Jackson is acquitted on all counts after five month trial. B: Jackson makes his first public appearance since the trial to accept eight records from the Guinness World Records in London, including Most Successful Entertainer of All Time. C: Jackson issues Thriller 25. D: Jackson dies in Los Angeles.
Movies:
Revert Network of
"Safavid dynasty".
Dots are editors and arrows represent reverts.
Always the latest reverting node pair is enlarged.
Revert Map of
"Israel and the apartheid analogy".
Each dot represents a pair of reverting editors.
The location of dots is determined by the number of edits already made
by the reverting/reverted editors on x/y axis.
Size of the dots is proportional to the number of reverts between the same pair.
Who we are:
Taha Yasseri, Robert Sumi, András Rung, János Török, András Kornai, János Kertész from Budapest University of Technology and Economis
and Hoda Sepehri Rad from University of Alberta.


Last modified: 25 May 2013, by Taha Yasseri.