In science news vol. 164 #25&26 the main article entitled “Bookish Math” is
about modern techniques for analyzing text, collectively called stylometry.
Using stylometry on various anonymously written or disputed texts has given
new data that has helped to resolve the true author. This may have impact on
future content based filters.
The full text of the article “Bookish Math” may be found at the science news
website
http://www.sciencenews.org
of particular interest to me are support-vector machines.
from the article:
“As does PCA[another technique mentioned], the new technique plots each
chunk of text as a point in a high-dimensional space. It then searches for
the best-fitting surface that divides the points belonging to one author
from those of the other author.”
I mentioned this idea before, but did not know this actually existed for
text classification.
http://article.gmane.org/gmane.ietf.asrg/6093
John Fenley
_________________________________________________________________
Make your home warm and cozy this winter with tips from MSN House & Home.
http://special.msn.com/home/warmhome.armx
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg