Revision as of 15:11, 7 April 2009

Part 1: project profile

Project name

Techniques to analyze the Wikipedia Social Network

Project short description

Wikipedia is the largest and most known example of wiki. Thousands of users cooperate within it to build a neutral encyclopedia. But not every user interacts equally: there are users who spend much more time on Wikipedia than others and may influence it. Whit its growing importance in the spread of information some questions are rising: how much the content of Wikipedia can be trusted? Who really writes on it? And what are his aims? Our work tries to study the community of users acting on a generic wiki with an automatic, general and modular process. The first step consists in the extraction of data from the MediaWiki dump. The second calculates the user contribution for each page of the wiki. Then those users who can be considered the "best" contributors are selected and used to build a Social Network. A standard set of analysis are applied to this network, such as macroscopic study and centrality measures.

Dates

Start date: 2008/07/01

End date: 2009/04/20

People involved

Project Advisor

David Laniado

Students

Students currently working on the project

Riccardo Tasso - rayman (ta) fastwebnet (tod) it

Part 2: project description

Related resources

• WikiTrust project

Software for Social Network Analysis

• Pajek: Program for Large Network Analysis
• JUNG Java Universal Network/Graph Framework
• The igraph library