import nltk

 from nltk.corpus import brown

 def common_suffixes_fun():

     suffix_fdist=nltk.FreqDist()

     for word in brown.words():

         word=word.lower()

         suffix_fdist[word[-1:]] +=1

         suffix_fdist[word[-2:]] +=1

         suffix_fdist[word[-3:]] +=1

     most_freqent_items=[it for it in sorted(suffix_fdist.items(),key=lambda x:(-x[1],x[0]))[:100]]

     return [su[0] for su in most_freqent_items]

 common_suffixes = common_suffixes_fun()

 def pos_features(word):

     features={}

     for su in common_suffixes:

         features['endswith(%s)' % su]=word.lower().endswith(su)

     return features

 def test_pos():

     tagged_words = brown.tagged_words(categories='news')[:5000]

     featuresets=[(pos_features(word),tag) for (word,tag) in tagged_words]

     size= int(len(tagged_words)*0.1)

     train_set, test_set = featuresets[size:],featuresets[:size]

     classifier=nltk.NaiveBayesClassifier.train(train_set)

     print nltk.classify.accuracy(classifier,test_set)

     classifier.show_most_informative_features(5)