Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with
or
.
Download ZIP
article quality visualization for english wikipedia
Python JavaScript
Branch: master

Fetching latest commit…

Cannot retrieve the latest commit at this time

Failed to load latest commit information.
article_html fix jquery, add more CLI arguments.
assets dashboard input in progress cell fills in yellow for retries
inputs added avg depth in dom; get param for input_server; various fixes for…
js Merge branch 'python-port' of github.com:slaporte/qualityvis
orange_reports/metamars_v1 first pass on metamars. 80+% accuracy on round-tripping article class…
orange_schemas
orange_scripts fix clean_missing_data
results
views Merge branch 'master' of github.com:slaporte/qualityvis
.gitignore new orange schemas
.json_exporter.py.swp resolved merge conflicts; classifier experiments
INTERESTING.md new note
LICENSE merge in stephen's new domStats and bing web stats
NOTES.md add a notes file with some CLI conversion notes
TODO.txt mean average error of around 6% :D
all_revisions.py
bottle_compressor.py
dashboard.py update exporter to remove tablib dependency, tweak dashboard/loupe
export_settings.py
exporter.py new version of MetaMars
input_server.py
loupe.py
loupe_list.py cli to create list of article titles
one_row.tmp.txt
progress.py python is happening
readme.markdown merge in stephen's new domStats and bing web stats
requirements.txt wrapping up requirements fixes
screen.png screenshot in markdown
stats.py add a couple more dom things, fix prefixing, update json_exporter, wh…
wapiti.py

readme.markdown

Article Quality Visualization

This is a javascript gadget for visualizing article quality in the English Wikipdia. We built this gadget for the San Francisco Mediawiki Hackathon in 2012.

Demo screenshot

Instillation and use

To use on the English Wikipedia, copy the contents of gadget.js into your common.js, and copy the contents of style.css into your common.css. For more detailed instructions on installing userscripts, see here.

Methodology

We used a number of metrics aggregated into four areas of quality: richness, structure, integratedness, community, citations, and significance. Each metric can be individually weighted, although they have not all been incorporated into the ranking formula.

Metrics

  • Assessment grade (featured article, good article, etc.)
  • Last 50 edits, grouped by editor
  • Category count
  • External links in the "External links" section
  • External links anywhere in the article
  • External links section count
  • Feedback score for completeness
  • Feedback score for objectivity
  • Feedback score for trustworthiness
  • Number of Google News results
  • Number of Google Web results
  • Image count
  • Incoming link count
  • Internal link count
  • Intro paragraph count
  • Page visits by date in the last month (according to http://stats.grok.se/)
  • Number of inline tags for POV statements
  • Number of inline tags for statements needing citation
  • Reference count
  • Reference section count
  • Likelihood the last revision was vandalism (according to http://www.wikitrust.net/)

About

This tool was built by:

  • Ben Plowman
  • Mahmoud Hashemi
  • Sarah Nahm
  • Stephen LaPorte

...with help and food from the awesome hosts of the hackathon!

Copyright 2012, licensed under the GPL 3.0. See LICENSE.

Something went wrong with that request. Please try again.