« Using Amazon Mechanical Turk to Extend Wikipedia Base | Main | We are extending deadlines for 'Web As A Stream' session at FQAS »

May 21, 2009


Feed You can follow this conversation by subscribing to the comment feed for this post.


Did you try to analyze community structure of Wikipedia category graph? That is, many category inclusions seem to be rather spurious, and perhaps hierarchical clustering of the category graph could help to detect such spurious inclusions.

Dmitry Lizorkin

We haven't analyzed Wikipedia category graph alone for detecting communities there, but this would indeed be the instructive thing to experiment with. And computationally it should be more manageable than for the whole set of Wikipedia articles :)
Thank you for your interesting suggestion!
We will think of performing this experiment and we'll let you know of the results if there be any interesting findings.

affordable banner printing

Interestingly the use of this material is one of the clothing I most closely associate with the Confederate flag as a poster for the race.

Dating agencies Review

But the second case does not actually mean that Wikipedia categories provide the "ground truth", while any other data organization is incorrect.

Social Kik

Its insane, that is what i can say concerning this post. Because this def is what the whole thing is about right? Keep it up!

how to find mr. right

I don't mind about this algorithm stuff. Using wikipedia as my close reference is still my main concern.

BOokkepper CAloundra

great post!wikipedia is really reliable source.

The comments to this entry are closed.

Become a Fan