Difference between revisions of "Wikipedia Category Graph"

From AIRWiki
Jump to: navigation, search
(aggiornati dopo filtraggio categorie non semantiche)
(Results of the analysis)
Line 30: Line 30:
 
After transferring the structure in a Neo4j graph is possible to create from it a Pajek file (.net) to make general analysis like described [[Social_Network_Analysis_With_Igraph_Package_Using_R|here]].
 
After transferring the structure in a Neo4j graph is possible to create from it a Pajek file (.net) to make general analysis like described [[Social_Network_Analysis_With_Igraph_Package_Using_R|here]].
  
===Results of the analysis===
+
==Results of the analysis==
 
We can use the tools described [http://igraph.sourceforge.net/doc/R/00Index.html here]
 
We can use the tools described [http://igraph.sourceforge.net/doc/R/00Index.html here]
  
Line 38: Line 38:
  
 
The '''graph density''' is 6.28*10<sup>-7</sup>. This is the ratio of the number of edges and the number of possible edges.
 
The '''graph density''' is 6.28*10<sup>-7</sup>. This is the ratio of the number of edges and the number of possible edges.
 +
 +
===Strongly connected components===
 +
 +
By applying the Tarjan's strongly connected components algorithm to the graph is possible to found 93 structures of up to 2 nodes. Each of them contains at least one cycle. Most of them are composed of two categories about the same thing, like ''History of the Germanic peoples'' and ''Ancient Germanic peoples'', but there are also more curious cases like this one
 +
[[Image:Struttura fortemente connessa wikipedia.png]]
 +
 +
=== Tested algorithms ===
 +
We tried 9 algorithms to choose the category which fit best an article. After confronting the results of the automatic procedure with human made assignments, the best algorithm was choose a different weight to each edge by the traversal direction.
 +
Sizes of the macrocategories determined this way are:
 +
[[Image:Dimensioni macrocategorie costi.png]]
  
 
==Download==
 
==Download==

Revision as of 11:18, 13 September 2010

Wikipedia Category Graph
Short Description: Represent Wikipedia Categories with a model based on graphs to further analyze it.
Coordinator: MarcoColombetti (colombet@elet.polimi.it)
Tutor: DavidLaniado (david.laniado@gmail.com), RiccardoTasso (tasso@elet.polimi.it)
Collaborator:
Students: JacopoFarina (jacopo1.farina@mail.polimi.it)
Research Area: Social Software and Semantic Web
Research Topic: Graph Mining and Analysis
Start: 2010/06/10
End: 2010/10/01
Status: Active
Level: Bs
Type: Course

The goal of the project is to analyze Wikipedia categories by representing them in a graph based database.

Wikipedia categories are not a three-based structure: a category may be contained in another one which is contained in another one which is contained in the first, generating a cyclic reference and many categories may be a root category (non contained in others).

For this reasons a graph database is better to represent the structure.

Creation and further analysis of the database with igraph

Wikipedia let users download the entire site database (with all versions of all articles) or just some selections of it. We use a selection which contains the category list and articles memberships in them.


Neo4j is a graph-based database, which allow a program to create and manipulate graph structures like nodes and relationships.

In order to transfer the database in neo4j format is better save it in a file, which will be read one line at time.


After transferring the structure in a Neo4j graph is possible to create from it a Pajek file (.net) to make general analysis like described here.

Results of the analysis

We can use the tools described here

The diameter of the graph is 32. This is the maximum distance (number of nodes in the minimum path) between two nodes. These two nodes are Prehistoric life sorted by geography (A category about prehistoric animals without articles) and BMW M20 (a car).

The average distance between two nodes is 5.5568.

The graph density is 6.28*10-7. This is the ratio of the number of edges and the number of possible edges.

Strongly connected components

By applying the Tarjan's strongly connected components algorithm to the graph is possible to found 93 structures of up to 2 nodes. Each of them contains at least one cycle. Most of them are composed of two categories about the same thing, like History of the Germanic peoples and Ancient Germanic peoples, but there are also more curious cases like this one Struttura fortemente connessa wikipedia.png

Tested algorithms

We tried 9 algorithms to choose the category which fit best an article. After confronting the results of the automatic procedure with human made assignments, the best algorithm was choose a different weight to each edge by the traversal direction. Sizes of the macrocategories determined this way are: Dimensioni macrocategorie costi.png

Download

  • The relation about the project (in Italian) can be found here
  • Sources in Java can be found here

Previous Work

What's in Wikipedia? Mapping Topics and Conflict Using Socially