Difference between revisions of "Vews: a Wikipedia Vandal Early Warning System"

From Wikipedia Quality
Jump to: navigation, search
(+ Embed)
(+ category)
 
Line 33: Line 33:
 
</nowiki>
 
</nowiki>
 
</code>
 
</code>
 +
 +
 +
 +
[[Category:Scientific works]]

Latest revision as of 06:22, 22 June 2020


Vews: a Wikipedia Vandal Early Warning System
Authors
Srijan Kumar
Francesca Spezzano
V. S. Subrahmanian
Publication date
2015
DOI
10.1145/2783258.2783367
Links
Original Preprint

Vews: a Wikipedia Vandal Early Warning System - scientific work related to Wikipedia quality published in 2015, written by Srijan Kumar, Francesca Spezzano and V. S. Subrahmanian.

Overview

Authors study the problem of detecting vandals on Wikipedia before any human or known vandalism detection system reports flagging potential vandals so that such users can be presented early to Wikipedia administrators. Authors leverage multiple classical ML approaches, but develop 3 novel sets of features. Authors Wikipedia Vandal Behavior (WVB) approach uses a novel set of user editing patterns as features to classify some users as vandals. Authors Wikipedia Transition Probability Matrix (WTPM) approach uses a set of features derived from a transition probability matrix and then reduces it via a neural net auto-encoder to classify some users as vandals. The VEWS approach merges the previous two approaches. Without using any information (e.g. reverts) provided by other users, these algorithms each have over 85% classification accuracy. Moreover, when temporal recency is considered, accuracy goes to almost 90%. Authors carry out detailed experiments on a new data set authors have created consisting of about 33K Wikipedia users (including both a black list and a white list of editors) and containing 770K edits. Authors describe specific behaviors that distinguish between vandals and non-vandals. Authors show that VEWS beats ClueBot NG and STiki, the best known algorithms today for vandalism detection. Moreover, VEWS detects far more vandals than ClueBot NG and on average, detects them 2.39 edits before ClueBot NG when both detect the vandal. However, authors show that the combination of VEWS and ClueBot NG can give a fully automated vandal early warning system with even higher accuracy.

Embed

Wikipedia Quality

Kumar, Srijan; Spezzano, Francesca; Subrahmanian, V. S.. (2015). "[[Vews: a Wikipedia Vandal Early Warning System]]".DOI: 10.1145/2783258.2783367.

English Wikipedia

{{cite journal |last1=Kumar |first1=Srijan |last2=Spezzano |first2=Francesca |last3=Subrahmanian |first3=V. S. |title=Vews: a Wikipedia Vandal Early Warning System |date=2015 |doi=10.1145/2783258.2783367 |url=https://wikipediaquality.com/wiki/Vews:_a_Wikipedia_Vandal_Early_Warning_System}}

HTML

Kumar, Srijan; Spezzano, Francesca; Subrahmanian, V. S.. (2015). &quot;<a href="https://wikipediaquality.com/wiki/Vews:_a_Wikipedia_Vandal_Early_Warning_System">Vews: a Wikipedia Vandal Early Warning System</a>&quot;.DOI: 10.1145/2783258.2783367.