Bayes classifier in NLTK,[url=http://www.pololacoste-pascher.fr/]polo lacoste[/url]
The features in the NLTK bayes classifier are "nominal",[url=http://www.herveleger.in.net/]herve leger[/url], not numeric. This means they can take a finite number of discrete values (labels),[url=http://www.jimmychooshoes-outlet.net/]jimmy choo outlet[/url], but they can't be treated as frequencies.
So with the Bayes classifier,[url=http://www.canadagoose.cc/]@#$%& parka[/url], you cannot directly use word frequency as a feature you could do something like use the 50 more frequent words from each text as your feature set,[url=http://www.montrepas-cher.fr/]montre fossil[/url], but that's quite a different thing
If your sentence has the same word multiple times,[url=http://www.@#$%&-rosheruns@#$%&/]roshe run hyperfuse[/url], it will just add the probs multiple times. If the word appears multiple times in the same class,[url=http://www.ray--ban.it/]rayban[/url], your training data should reflect that in the word count.
For added accuracy,[url=http://www.chiflatiron.in.net/]chi hair straightener[/url], count all bi grams,[url=http://www.longchamphandbags-outlet@#$%&/]longchamp bags[/url], tri grams,[url=http://www.nikeblazer-pascher.fr/]blazer @#$%&[/url], etc as separate features.
It helps to manually write your own classifiers so that you understand exactly what is happening and what you need to do to imporve accuracy. If you use a pre packaged solution and it doesn't work well enough,[url=http://www.christianlouboutinshoes.in.net/]louboutin sale[/url], there is not much you can do about it.

[url=http://bbs.cdscjd@#$%&/forum.php?mod=viewthread&tid=5001611]burberry @#$%& Alleged LAX Shooter Reportedly Concealed Weapon By Cutting Hol[/url]

[url=http://www.ocvo.fr/galerie/picture.php?/247/list/1562,1425,2258,1585,1292,2630,3554,3080,2087,1256,1510,1285,2899,3788,962,1173,2271,1850,322,852,1675,830,247,1667,607/start-20&comments_order=DESC#comments]mac make up and Treatment[/url]

[url=http://www.33mscx33msc@#$%&/forum.php?mod=viewthread&tid=4148146]babyliss pro nano titanium AIRBAGS TRIGGER NEW FEARS[/url]

[url=http://www.shuizuwanjia@#$%&/forum.php?mod=viewthread&tid=945215&fromuid=53761]christian louboutin Achieving Full Golf Swing Success[/url]

[url=http://people.shangdu@#$%&/home.php?mod=space&uid=157396&do=blog&quickforward=1&id=833076]louis @#$%& factory outlet 10 Reuses for Single[/url]