• View
• Edit
• History
• Print

Statistics

Back to Research

Objective

• Using statistics as a strategical tool that every user of Seedea can leverage in order to innovate more efficiently and pleasantly.

General question

1. So what kind of interesting statistics can we do an a wiki?
2. What is a wiki mathematically speaking?
3. What are statistics good for?

Functional modules

1. gather data from an existing network
2. analyze data based on relevant statistical analysis tools and output the result
3. detect patterns and add suggestions to the result based on good practices
4. present visually the result in order to facilitate decision making

Dataset

• Content of this very wiki : seedea.org
• Possibility to extract and prepare the dataset
  • calling pmWiki API
  • executing CGI scripts locally to directly scrap data from the filesystem (based on PmWiki/PageFileFormat)
  • making a PHP module for pmWiki that will generate data to parse eventually from a remote server

Going further

• Idea life-line : Quickly visually seeing how an idea is evolving.
• Visual ant wiki : Draw a skinnable sitemap of an existing wiki based on graph theory.
• Computer Assisted Self-Teaching alias CAST : Use the computer to do meta-analysis on your current task to provide suggestions for improvement based on community good practices.

References

• Getting Better Statistics about Your Wiki by Jon Silvers for The Atlassian Blog
• VisualComplexity.com : A visual exploration on mapping complex networks
• Visualizing Data By Ben Fry for O'Reilly Media
• theinfo : for people with large data sets
• WikiR Structure : Integrate R language into MediaWiki

To explore

• Wikimetrics deals with measurement of Wikis. That includes specifying metrics and monitoring the frequency and consistency of their measurement, and their usefulness to users and researchers.
• Data Analysis Using Regression and Multilevel/Hierarchical Models by Gelman and Hill (including R/Bugs code)