During my work with natural language processing related projects I was involved in development of several classifiers for production and research.
EXAMPLE: http://korobov-labs.com/classifier/
SVM Classifiers (scikit-learn, svm-light as well as my own SVM implementation)
Naive Bayesian Classifiers (scikit-learn, as well as my own NBC implementation)
I shared a classifier working instance (it is used for research activities ; it has slightly trained model ; it is dynamically retrained periodically by a script, so it could be irresponsible for short periods, but usually it is available)
Also you can try XML-RPC endpoint for this service by executing this Python 2.7 code:
from xmlrpclib import ServerProxy
connect = ServerProxy("http://korobov-labs.com:8001")
print connect.classify_text('Put your English text here...')
or for a more convenient way to try it on a real text, try this one :
from xmlrpclib import ServerProxy
import urllib2
connect = ServerProxy("http://korobov-labs.com:8001")
# Get sherlock holmes text from the internet
data = urllib2.urlopen(
'http://sherlock-holm.es/stories/plain-text/twis.txt').read()
print connect.classify_text(data)