ML Wiki
Machine Learning Wiki - A collection of ML concepts, algorithms, and resources.

Controversy Detection

Controversy Detection

Cotroversy Detection

Controversy as defined in the previous chapter is a state of prolonged public dispute or debate, usually concerning a matter of conflicting opinion or point of view.

Examples

  • “global warming is a hoax”
  • political debates, economic issues
  • same-sex marriage
  • healthcare

Results: not trustworthy information, e.g. “vaccines cause autism”, etc

correlation: if a wikipedia page is controversial so is the topic of the page

Features:

  • len of page
  • proportion of anonimous edits
  • number of reverts

hierarchy of controversial words and topics: [Awadallah2012, Popescu2010]

sentiment can also be a useful feature for controversy detection

Approach:

  • map webpage to a wiki article
  • if the wiki article is controversial, so is the webpage

mapping:

  • querying the wikipedia api
  • take top 10 terms (e.g. most frequent expect stopwords) and then run a commercial search engine on them - and find the wiki page in the results

Wikipedia scores:

  • D - if dispute tags are absent - see Kittur2017, Rad2012. usually added to the “Talk” page, but only a small fraction of pages have them
  • C - predict controversy using wiki page metadata: see Kittur2017, model: Das2013
  • len of the page and its talk page
  • number of editors, number of anonimous editors
  • M - based on the concept of manual reverts and edit wars on wikipedia
  • based on the reputation of the users involved in the “wars”

These 3 scores are combined into one. D is omitted because the coverage is too small to have any effect on the model

Data: http://ciir.cs.umass.edu/download

Papers:

  • Awadallah2012, Harmony and dissonance: organizing the people’s voices on political controversies [http://dl.acm.org/citation.cfm?id=2124359&dl=ACM&coll=DL&CFID=685632707&CFTOKEN=42534054]
  • Das2013 Manipulation Among the Arbiters of Collective Intelligence: How Wikipedia Administrators Mold Public Opinion [http://www.cse.wustl.edu/~allenlavoie/papers/wiki-cikm.pdf]
  • Kittur2017, He says, she says: conflict and coordination in Wikipedia [http://www.kittur.org/files/Kittur_2007_Wikipedia_CHI.pdf]
  • Popescu2010, Detecting controversial events from twitter [http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.458.1082&rep=rep1&type=pdf]
  • Rad2012, Identifying controversial articles in Wikipedia: A comparative study, [http://www.opensym.org/ws2012/p18wikisym2012.pdf]