Size matters: word count as a measure of quality on Wikipedia

From Brede Wiki
Jump to: navigation, search
Conference paper (help)
Size matters: word count as a measure of quality on Wikipedia
Authors: Joshua E. Blumenstock
Citation: Proceeding of the 17th international conference on World Wide Web  : 2008
Editors:
Publisher: Define publisher
Meeting: 17th international conference on World Wide Web
Database(s): AcaWiki Citeulike
DOI: 10.1145/1367497.1367673.
Link(s): http://wwwconference.org/www2008/papers/pdf/p1095-jblumenstock.pdf
Search
Web: DuckDuckGo Bing Google Yahoo!Google PDF
Article: Google Scholar PubMed
Restricted: DTU Digital Library
Services
Format: BibTeX

Size matters: word count as a measure of quality on Wikipedia is a study on quality of Wikipedia articles. The author tries a number of different features for predicting the quality and finds that word count is a simple and good predictor when classifying articles whether they are featured or randomly selected.

Contents

[edit] Methods

[edit] Features

  1. Word count
  2. Frequency counts
    1. Character count
    2. ...
    3. Readability indicies
    4. ...
  3. Structural features
    1. Internal links
    2. Category count
    3. Citation count
    4. ...

[edit] Models

  1. Cut at 2000 words
  2. Multi-layer perceptron
  3. K-nearest neighbor classifier
  4. Logit model
  5. Random-forest classifier.

[edit] Data

English Wikipedia articles with:

  • 1554 featured articles
  • 9513 randomly selected.

[edit] Related papers

  1. A multimethod study of information quality in wiki collaboration
  2. Assessing information quality of a community-based encyclopedia
  3. Computing trust from revision history
Personal tools