It currently demonstrates the following functionality:

• part-of-speech tagging with the default NLTK pos tagger
• chunking and named entity recognition with the default NLTK chunker
• sentiment analysis with a combination of a naive bayes classifier and a maximum entropy classifier, both trained on the movie reviews corpus

If you like it, please share it. If you want to see more, leave a comment below. And if you are interested in a service that could apply these processes to your own data, please fill out this NLTK services survey.

Other Natural Language Processing Demos

Here's a list of similar resources on the web:

• A demo of the Stanford Parser with a javascript API: Natural-language Parsing For The Web
• A demo of the FreeLing language analysis suite: FreeLing Demo
• Emotional identification from text: EmoLib

Eliminating low information features gives your model clarity by removing noisy data. It can save you from overfitting and the curse of dimensionality. When you use only the higher information features, you can increase performance while also decreasing the size of the model, which results in less memory usage along with faster training and classification. Removing features may seem intuitively wrong, but wait till you see the results.

High Information Feature Selection

Using the same evaluate_classifier method as in the previous post on classifying with bigrams, I got the following results using the 10000 most informative words:

evaluating best word features
accuracy: 0.93
pos precision: 0.890909090909
pos recall: 0.98
neg precision: 0.977777777778
neg recall: 0.88
Most Informative Features
             magnificent = True              pos : neg    =     15.0 : 1.0
             outstanding = True              pos : neg    =     13.6 : 1.0
               insulting = True              neg : pos    =     13.0 : 1.0
              vulnerable = True              pos : neg    =     12.3 : 1.0
               ludicrous = True              neg : pos    =     11.8 : 1.0
                  avoids = True              pos : neg    =     11.7 : 1.0
             uninvolving = True              neg : pos    =     11.7 : 1.0
              astounding = True              pos : neg    =     10.3 : 1.0
             fascination = True              pos : neg    =     10.3 : 1.0
                 idiotic = True              neg : pos    =      9.8 : 1.0

Contrast this with the results from the first article on classification for sentiment analysis, where we use all the words as features:

evaluating single word features
accuracy: 0.728
pos precision: 0.651595744681
pos recall: 0.98
neg precision: 0.959677419355
neg recall: 0.476
Most Informative Features
         magnificent = True              pos : neg    =     15.0 : 1.0
         outstanding = True              pos : neg    =     13.6 : 1.0
           insulting = True              neg : pos    =     13.0 : 1.0
          vulnerable = True              pos : neg    =     12.3 : 1.0
           ludicrous = True              neg : pos    =     11.8 : 1.0
              avoids = True              pos : neg    =     11.7 : 1.0
         uninvolving = True              neg : pos    =     11.7 : 1.0
          astounding = True              pos : neg    =     10.3 : 1.0
         fascination = True              pos : neg    =     10.3 : 1.0
             idiotic = True              neg : pos    =      9.8 : 1.0

The accuracy is over 20% higher when using only the best 10000 words and pos precision has increased almost 24% while neg recall improved over 40%. These are huge increases with no reduction in pos recall and even a slight increase in neg precision. Here's the full code I used to get these results, with an explanation below.

import collections, itertools
import nltk.classify.util, nltk.metrics
from nltk.classify import NaiveBayesClassifier
from nltk.corpus import movie_reviews, stopwords
from nltk.collocations import BigramCollocationFinder
from nltk.metrics import BigramAssocMeasures
from nltk.probability import FreqDist, ConditionalFreqDist

def evaluate_classifier(featx):
	negids = movie_reviews.fileids('neg')
	posids = movie_reviews.fileids('pos')

	negfeats = [(featx(movie_reviews.words(fileids=[f])), 'neg') for f in negids]
	posfeats = [(featx(movie_reviews.words(fileids=[f])), 'pos') for f in posids]

	negcutoff = len(negfeats)*3/4
	poscutoff = len(posfeats)*3/4

	trainfeats = negfeats[:negcutoff] + posfeats[:poscutoff]
	testfeats = negfeats[negcutoff:] + posfeats[poscutoff:]

	classifier = NaiveBayesClassifier.train(trainfeats)
	refsets = collections.defaultdict(set)
	testsets = collections.defaultdict(set)

	for i, (feats, label) in enumerate(testfeats):
			refsets[label].add(i)
			observed = classifier.classify(feats)
			testsets[observed].add(i)

	print 'accuracy:', nltk.classify.util.accuracy(classifier, testfeats)
	print 'pos precision:', nltk.metrics.precision(refsets['pos'], testsets['pos'])
	print 'pos recall:', nltk.metrics.recall(refsets['pos'], testsets['pos'])
	print 'neg precision:', nltk.metrics.precision(refsets['neg'], testsets['neg'])
	print 'neg recall:', nltk.metrics.recall(refsets['neg'], testsets['neg'])
	classifier.show_most_informative_features()

def word_feats(words):
	return dict([(word, True) for word in words])

print 'evaluating single word features'
evaluate_classifier(word_feats)

word_fd = FreqDist()
label_word_fd = ConditionalFreqDist()

for word in movie_reviews.words(categories=['pos']):
	word_fd.inc(word.lower())
	label_word_fd['pos'].inc(word.lower())

for word in movie_reviews.words(categories=['neg']):
	word_fd.inc(word.lower())
	label_word_fd['neg'].inc(word.lower())

# n_ii = label_word_fd[label][word]
# n_ix = word_fd[word]
# n_xi = label_word_fd[label].N()
# n_xx = label_word_fd.N()

pos_word_count = label_word_fd['pos'].N()
neg_word_count = label_word_fd['neg'].N()
total_word_count = pos_word_count + neg_word_count

word_scores = {}

for word, freq in word_fd.iteritems():
	pos_score = BigramAssocMeasures.chi_sq(label_word_fd['pos'][word],
		(freq, pos_word_count), total_word_count)
	neg_score = BigramAssocMeasures.chi_sq(label_word_fd['neg'][word],
		(freq, neg_word_count), total_word_count)
	word_scores[word] = pos_score + neg_score

best = sorted(word_scores.iteritems(), key=lambda (w,s): s, reverse=True)[:10000]
bestwords = set([w for w, s in best])

def best_word_feats(words):
	return dict([(word, True) for word in words if word in bestwords])

print 'evaluating best word features'
evaluate_classifier(best_word_feats)

def best_bigram_word_feats(words, score_fn=BigramAssocMeasures.chi_sq, n=200):
	bigram_finder = BigramCollocationFinder.from_words(words)
	bigrams = bigram_finder.nbest(score_fn, n)
	d = dict([(bigram, True) for bigram in bigrams])
	d.update(best_word_feats(words))
	return d

print 'evaluating best words + bigram chi_sq word features'
evaluate_classifier(best_bigram_word_feats)

Calculating Information Gain

To find the highest information features, we need to calculate information gain for each word. Information gain for classification is a measure of how common a feature is in a particular class compared to how common it is in all other classes. A word that occurs primarily in positive movie reviews and rarely in negative reviews is high information. For example, the presence of the word "magnificent" in a movie review is a strong indicator that the review is positive. That makes "magnificent" a high information word. Notice that the most informative features above did not change. That makes sense because the point is to use only the most informative features and ignore the rest.

One of the best metrics for information gain is chi square. NLTK includes this in the BigramAssocMeasures class in the metrics package. To use it, first we need to calculate a few frequencies for each word: its overall frequency and its frequency within each class. This is done with a FreqDist for overall frequency of words, and a ConditionalFreqDist where the conditions are the class labels. Once we have those numbers, we can score words with the BigramAssocMeasures.chi_sq function, then sort the words by score and take the top 10000. We then put these words into a set, and use a set membership test in our feature selection function to select only those words that appear in the set. Now each file is classified based on the presence of these high information words.

Signficant Bigrams

The code above also evaluates the inclusion of 200 significant bigram collocations. Here are the results:

evaluating best words + bigram chi_sq word features
accuracy: 0.92
pos precision: 0.913385826772
pos recall: 0.928
neg precision: 0.926829268293
neg recall: 0.912
Most Informative Features
             magnificent = True              pos : neg    =     15.0 : 1.0
             outstanding = True              pos : neg    =     13.6 : 1.0
               insulting = True              neg : pos    =     13.0 : 1.0
              vulnerable = True              pos : neg    =     12.3 : 1.0
       ('matt', 'damon') = True              pos : neg    =     12.3 : 1.0
          ('give', 'us') = True              neg : pos    =     12.3 : 1.0
               ludicrous = True              neg : pos    =     11.8 : 1.0
             uninvolving = True              neg : pos    =     11.7 : 1.0
                  avoids = True              pos : neg    =     11.7 : 1.0
    ('absolutely', 'no') = True              neg : pos    =     10.6 : 1.0

This shows that bigrams don't matter much when using only high information words. In this case, the best way to evaluate the difference between including bigrams or not is to look at precision and recall. With the bigrams, you we get more uniform performance in each class. Without bigrams, precision and recall are less balanced. But the differences may depend on your particular data, so don't assume these observations are always true.

Improving Feature Selection

The big lesson here is that improving feature selection will improve your classifier. Reducing dimensionality is one of the single best things you can do to improve classifier performance. It's ok to throw away data if that data is not adding value. And it's especially recommended when that data is actually making your model worse.

1. filter out stopwords
2. include bigram collocations

To do this effectively, we'll modify the previous code so that we can use an arbitrary feature extractor function that takes the words in a file and returns the feature dictionary. As before, we'll use these features to train a Naive Bayes Classifier.

import collections
import nltk.classify.util, nltk.metrics
from nltk.classify import NaiveBayesClassifier
from nltk.corpus import movie_reviews

def evaluate_classifier(featx):
	negids = movie_reviews.fileids('neg')
	posids = movie_reviews.fileids('pos')

	negfeats = [(featx(movie_reviews.words(fileids=[f])), 'neg') for f in negids]
	posfeats = [(featx(movie_reviews.words(fileids=[f])), 'pos') for f in posids]

	negcutoff = len(negfeats)*3/4
	poscutoff = len(posfeats)*3/4

	trainfeats = negfeats[:negcutoff] + posfeats[:poscutoff]
	testfeats = negfeats[negcutoff:] + posfeats[poscutoff:]

	classifier = NaiveBayesClassifier.train(trainfeats)
	refsets = collections.defaultdict(set)
	testsets = collections.defaultdict(set)

	for i, (feats, label) in enumerate(testfeats):
			refsets[label].add(i)
			observed = classifier.classify(feats)
			testsets[observed].add(i)

	print 'accuracy:', nltk.classify.util.accuracy(classifier, testfeats)
	print 'pos precision:', nltk.metrics.precision(refsets['pos'], testsets['pos'])
	print 'pos recall:', nltk.metrics.recall(refsets['pos'], testsets['pos'])
	print 'neg precision:', nltk.metrics.precision(refsets['neg'], testsets['neg'])
	print 'neg recall:', nltk.metrics.recall(refsets['neg'], testsets['neg'])
	classifier.show_most_informative_features()

Baseline Bag of Words Feature Extraction

Here's the baseline feature extractor for bag of words feature selection.

def word_feats(words):
	return dict([(word, True) for word in words])

evaluate_classifier(word_feats)

The results are the same as in the previous articles, but I've included them here for reference:

accuracy: 0.728
pos precision: 0.651595744681
pos recall: 0.98
neg precision: 0.959677419355
neg recall: 0.476
Most Informative Features
         magnificent = True              pos : neg    =     15.0 : 1.0
         outstanding = True              pos : neg    =     13.6 : 1.0
           insulting = True              neg : pos    =     13.0 : 1.0
          vulnerable = True              pos : neg    =     12.3 : 1.0
           ludicrous = True              neg : pos    =     11.8 : 1.0
              avoids = True              pos : neg    =     11.7 : 1.0
         uninvolving = True              neg : pos    =     11.7 : 1.0
          astounding = True              pos : neg    =     10.3 : 1.0
         fascination = True              pos : neg    =     10.3 : 1.0
             idiotic = True              neg : pos    =      9.8 : 1.0

Stopword Filtering

Stopwords are words that are generally considered useless. Most search engines ignore these words because they are so common that including them would greatly increase the size of the index without improving precision or recall. NLTK comes with a stopwords corpus that includes a list of 128 english stopwords. Let's see what happens when we filter out these words.

from nltk.corpus import stopwords
stopset = set(stopwords.words('english'))

def stopword_filtered_word_feats(words):
	return dict([(word, True) for word in words if word not in stopset])

evaluate_classifier(stopword_filtered_word_feats)

And the results for a stopword filtered bag of words are:

accuracy: 0.726
pos precision: 0.649867374005
pos recall: 0.98
neg precision: 0.959349593496
neg recall: 0.472

Accuracy went down .2%, and pos precision and neg recall dropped as well! Apparently stopwords add information to sentiment analysis classification. I did not include the most informative features since they did not change.

Bigram Collocations

As mentioned at the end of the article on precision and recall, it's possible that including bigrams will improve classification accuracy. The hypothesis is that people say things like "not great", which is a negative expression that the bag of words model could interpret as positive since it sees "great" as a separate word.

To find significant bigrams, we can use nltk.collocations.BigramCollocationFinder along with nltk.metrics.BigramAssocMeasures. The BigramCollocationFinder maintains 2 internal FreqDists, one for individual word frequencies, another for bigram frequencies. Once it has these frequency distributions, it can score individual bigrams using a scoring function provided by BigramAssocMeasures, such chi-square. These scoring functions measure the collocation correlation of 2 words, basically whether the bigram occurs about as frequently as each individual word.

import itertools
from nltk.collocations import BigramCollocationFinder
from nltk.metrics import BigramAssocMeasures

def bigram_word_feats(words, score_fn=BigramAssocMeasures.chi_sq, n=200):
	bigram_finder = BigramCollocationFinder.from_words(words)
	bigrams = bigram_finder.nbest(score_fn, n)
	return dict([(ngram, True) for ngram in itertools.chain(words, bigrams)])

evaluate_classifier(bigram_word_feats)

After some experimentation, I found that using the 200 best bigrams from each file produced great results:

accuracy: 0.816
pos precision: 0.753205128205
pos recall: 0.94
neg precision: 0.920212765957
neg recall: 0.692
Most Informative Features
         magnificent = True              pos : neg    =     15.0 : 1.0
         outstanding = True              pos : neg    =     13.6 : 1.0
           insulting = True              neg : pos    =     13.0 : 1.0
          vulnerable = True              pos : neg    =     12.3 : 1.0
   ('matt', 'damon') = True              pos : neg    =     12.3 : 1.0
      ('give', 'us') = True              neg : pos    =     12.3 : 1.0
           ludicrous = True              neg : pos    =     11.8 : 1.0
         uninvolving = True              neg : pos    =     11.7 : 1.0
              avoids = True              pos : neg    =     11.7 : 1.0
('absolutely', 'no') = True              neg : pos    =     10.6 : 1.0

Yes, you read that right, Matt Damon is apparently one of the best predictors for positive sentiment in movie reviews. But despite this chuckle-worthy result

• accuracy is up almost 9%
• pos precision has increased over 10% with only 4% drop in recall
• neg recall has increased over 21% with just under 4% drop in precision

So it appears that the bigram hypothesis is correct, and including significant bigrams can increase classifier effectiveness. Note that it's significant bigrams that enhance effectiveness. I tried using nltk.util.bigrams to include all bigrams, and the results were only a few points above baseline. This points to the idea that including only significant features can improve accuracy compared to using all features. In a future article, I'll try trimming down the single word features to only include significant words.

Classifier Precision

Precision measures the exactness of a classifier. A higher precision means less false positives, while a lower precision means more false positives. This is often at odds with recall, as an easy way to improve precision is to decrease recall.

Classifier Recall

Recall measures the completeness, or sensitivity, of a classifier. Higher recall means less false negatives, while lower recall means more false negatives. Improving recall can often decrease precision because it gets increasingly harder to be precise as the sample space increases.

F-measure Metric

Precision and recall can be combined to produce a single metric known as F-measure, which is the weighted harmonic mean of precision and recall. I find F-measure to be about as useful as accuracy. Or in other words, compared to precision & recall, F-measure is mostly useless, as you'll see below.

Measuring Precision and Recall of a Naive Bayes Classifier

The NLTK metrics module provides functions for calculating all three metrics mentioned above. But to do so, you need to build 2 sets for each classification label: a reference set of correct values, and a test set of observed values. Below is a modified version of the code from the previous article, where we trained a Naive Bayes Classifier. This time, instead of measuring accuracy, we'll collect reference values and observed values for each label (pos or neg), then use those sets to calculate the precision, recall, and F-measure of the naive bayes classifier. The actual values collected are simply the index of each featureset using enumerate.

import collections
import nltk.metrics
from nltk.classify import NaiveBayesClassifier
from nltk.corpus import movie_reviews

def word_feats(words):
	return dict([(word, True) for word in words])

negids = movie_reviews.fileids('neg')
posids = movie_reviews.fileids('pos')

negfeats = [(word_feats(movie_reviews.words(fileids=[f])), 'neg') for f in negids]
posfeats = [(word_feats(movie_reviews.words(fileids=[f])), 'pos') for f in posids]

negcutoff = len(negfeats)*3/4
poscutoff = len(posfeats)*3/4

trainfeats = negfeats[:negcutoff] + posfeats[:poscutoff]
testfeats = negfeats[negcutoff:] + posfeats[poscutoff:]
print 'train on %d instances, test on %d instances' % (len(trainfeats), len(testfeats))

classifier = NaiveBayesClassifier.train(trainfeats)
refsets = collections.defaultdict(set)
testsets = collections.defaultdict(set)

for i, (feats, label) in enumerate(testfeats):
	refsets[label].add(i)
	observed = classifier.classify(feats)
	testsets[observed].add(i)

print 'pos precision:', nltk.metrics.precision(refsets['pos'], testsets['pos'])
print 'pos recall:', nltk.metrics.recall(refsets['pos'], testsets['pos'])
print 'pos F-measure:', nltk.metrics.f_measure(refsets['pos'], testsets['pos'])
print 'neg precision:', nltk.metrics.precision(refsets['neg'], testsets['neg'])
print 'neg recall:', nltk.metrics.recall(refsets['neg'], testsets['neg'])
print 'neg F-measure:', nltk.metrics.f_measure(refsets['neg'], testsets['neg'])

Precision and Recall for Positive and Negative Reviews

I found the results quite interesting:

pos precision: 0.651595744681
pos recall: 0.98
pos F-measure: 0.782747603834
neg precision: 0.959677419355
neg recall: 0.476
neg F-measure: 0.636363636364

So what does this mean?

1. Nearly every file that is pos is correctly identified as such, with 98% recall. This means very few false negatives in the pos class.
2. But, a file given a pos classification is only 65% likely to be correct. Not so good precision leads to 35% false positives for the pos label.
3. Any file that is identified as neg is 96% likely to be correct (high precision). This means very few false positives for the neg class.
4. But many files that are neg are incorrectly classified. Low recall causes 52% false negatives for the neg label.
5. F-measure provides no useful information. There's no insight to be gained from having it, and we wouldn't lose any knowledge if it was taken away.

Improving Results with Better Feature Selection

One possible explanation for the above results is that people use normally positives words in negative reviews, but the word is preceded by "not" (or some other negative word), such as "not great". And since the classifier uses the bag of words model, which assumes every word is independent, it cannot learn that "not great" is a negative. If this is the case, then these metrics should improve if we also train on multiple words, a topic I'll explore in a future article.

Another possibility is the abundance of naturally neutral words, the kind of words that are devoid of sentiment. But the classifier treats all words the same, and has to assign each word to either pos or neg. So maybe otherwise neutral or meaningless words are being placed in the pos class because the classifier doesn't know what else to do. If this is the case, then the metrics should improve if we eliminate the neutral or meaningless words from the featuresets, and only classify using sentiment rich words. This is usually done using the concept of information gain, aka mutual information, to improve feature selection, which I'll also explore in a future article.

If you have your own theories to explain the results, or ideas on how to improve precision and recall, please share in the comments.

NLTK Naive Bayes Classification

NLTK comes with all the pieces you need to get started on sentiment analysis: a movie reviews corpus with reviews categorized into pos and neg categories, and a number of trainable classifiers. We'll start with a simple NaiveBayesClassifier as a baseline, using boolean word feature extraction.

Bag of Words Feature Extraction

All of the NLTK classifiers work with featstructs, which can be simple dictionaries mapping a feature name to a feature value. For text, we'll use a simplified bag of words model where every word is feature name with a value of True. Here's the feature extraction method:

def word_feats(words):
		return dict([(word, True) for word in words])

Training Set vs Test Set and Accuracy

The movie reviews corpus has 1000 positive files and 1000 negative files. We'll use 3/4 of them as the training set, and the rest as the test set. This gives us 1500 training instances and 500 test instances. The classifier training method expects to be given a list of tokens in the form of [(feats, label)] where feats is a feature dictionary and label is the classification label. In our case, feats will be of the form {word: True} and label will be one of 'pos' or 'neg'. For accuracy evaluation, we can use nltk.classify.util.accuracy with the test set as the gold standard.

Training and Testing the Naive Bayes Classifier

Here's the complete python code for training and testing a Naive Bayes Classifier on the movie review corpus.

import nltk.classify.util
from nltk.classify import NaiveBayesClassifier
from nltk.corpus import movie_reviews

def word_feats(words):
	return dict([(word, True) for word in words])

negids = movie_reviews.fileids('neg')
posids = movie_reviews.fileids('pos')

negfeats = [(word_feats(movie_reviews.words(fileids=[f])), 'neg') for f in negids]
posfeats = [(word_feats(movie_reviews.words(fileids=[f])), 'pos') for f in posids]

negcutoff = len(negfeats)*3/4
poscutoff = len(posfeats)*3/4

trainfeats = negfeats[:negcutoff] + posfeats[:poscutoff]
testfeats = negfeats[negcutoff:] + posfeats[poscutoff:]
print 'train on %d instances, test on %d instances' % (len(trainfeats), len(testfeats))

classifier = NaiveBayesClassifier.train(trainfeats)
print 'accuracy:', nltk.classify.util.accuracy(classifier, testfeats)
classifier.show_most_informative_features()

And the output is:

train on 1500 instances, test on 500 instances
accuracy: 0.728
Most Informative Features
         magnificent = True              pos : neg    =     15.0 : 1.0
         outstanding = True              pos : neg    =     13.6 : 1.0
           insulting = True              neg : pos    =     13.0 : 1.0
          vulnerable = True              pos : neg    =     12.3 : 1.0
           ludicrous = True              neg : pos    =     11.8 : 1.0
              avoids = True              pos : neg    =     11.7 : 1.0
         uninvolving = True              neg : pos    =     11.7 : 1.0
          astounding = True              pos : neg    =     10.3 : 1.0
         fascination = True              pos : neg    =     10.3 : 1.0
             idiotic = True              neg : pos    =      9.8 : 1.0

As you can see, the 10 most informative features are, for the most part, highly descriptive adjectives. The only 2 words that seem a bit odd are "vulnerable" and "avoids". Perhaps these words refer to important plot points or character development that signify a good movie. Whatever the case, with simple assumptions and very little code we're able to get almost 73% accuracy. This is somewhat near human accuracy, as apparently people agree on sentiment only around 80% of the time. Future articles in this series will cover precision & recall metrics, alternative classifiers, and techniques for improving accuracy.

• What’s the Difference Between Stemming and Lemmatization? - Ask Dr. Search
• A List of Social Tagging Datasets Made Available for Research
• Social Signaling and Language Use
• Lexical Growth in the Blogosphere
• Spelling correction using the Python Natural Language Toolkit (nltk)
• OPUS - an open source parallel corpus
• Evaluating POS Taggers: The Contenders
• Text Analytics Wiki

With the latest 2.0 beta releases (2.0b8 as of this writing), NLTK has included a ClassifierBasedTagger as well as a pre-trained tagger used by the nltk.tag.pos_tag method. Based on the name, then pre-trained tagger appears to be a ClassifierBasedTagger trained on the treebank corpus using a MaxentClassifier. So let's see how a classifier tagger compares to the brill tagger.

Training Sets

For the brown corpus, I trained on 2/3 of the reviews, lore, and romance categories, and tested against the remaining 1/3. For conll2000, I used the standard train.txt vs test.txt. And for treebank, I again used a 2/3 vs 1/3 split.

import itertools
from nltk.corpus import brown, conll2000, treebank

brown_reviews = brown.tagged_sents(categories=['reviews'])
brown_reviews_cutoff = len(brown_reviews) * 2 / 3
brown_lore = brown.tagged_sents(categories=['lore'])
brown_lore_cutoff = len(brown_lore) * 2 / 3
brown_romance = brown.tagged_sents(categories=['romance'])
brown_romance_cutoff = len(brown_romance) * 2 / 3

brown_train = list(itertools.chain(brown_reviews[:brown_reviews_cutoff],
	brown_lore[:brown_lore_cutoff], brown_romance[:brown_romance_cutoff]))
brown_test = list(itertools.chain(brown_reviews[brown_reviews_cutoff:],
	brown_lore[brown_lore_cutoff:], brown_romance[brown_romance_cutoff:]))

conll_train = conll2000.tagged_sents('train.txt')
conll_test = conll2000.tagged_sents('test.txt')

treebank_cutoff = len(treebank.tagged_sents()) * 2 / 3
treebank_train = treebank.tagged_sents()[:treebank_cutoff]
treebank_test = treebank.tagged_sents()[treebank_cutoff:]

Classifier Taggers

There are 3 new taggers referenced below:

• cpos is an instance of ClassifierBasedPOSTagger using the default NaiveBayesClassifier. It was trained by doing ClassifierBasedPOSTagger(train=train_sents)
• craubt is like cpos, but has the raubt tagger from part 2 as a backoff tagger by doing ClassifierBasedPOSTagger(train=train_sents, backoff=raubt)
• bcpos is a BrillTagger using cpos as its initial tagger instead of raubt.

The raubt tagger is the same as from part 2, and braubt is from part 3.

postag is NLTK's pre-trained tagger used by the pos_tag function. It can be loaded using nltk.data.load(nltk.tag._POS_TAGGER).

Accuracy Evaluation

Tagger accuracy was determined by calling the evaluate method with the test set on each trained tagger. Here are the results:

Conclusions

The above results are quite interesting, and lead to a few conclusions:

1. Training data is hugely significant when it comes to accuracy. This is why postag takes a huge nose dive on brown, while at the same time can get near 100% accuracy on treebank.
2. A ClassifierBasedPOSTagger does not need a backoff tagger, since cpos accuracy is exactly the same as for craubt across all corpora.
3. The ClassifierBasedPOSTagger is not necessarily more accurate than the bcraubt tagger from part 3 (at least with the default feature detector). It also takes much longer to train and tag (more details below) and so may not be worth the tradeoff in efficiency.
4. Using BrillTagger will nearly always increase the accuracy of your initial tagger, but not by much.

I was also surprised at how much more accurate postag was compared to cpos. Thinking that postag was probably trained on the full treebank corpus, I did the same, and re-evaluated:

cpos = ClassifierBasedPOSTagger(train=treebank.tagged_sents())
cpos.evaluate(treebank_test)

The result was 98.08% accuracy. So the remaining 2% difference must be due to the MaxentClassifier being more accurate than NaiveBayesClassifier, and/or the use of a different feature detector. I tried again with classifier_builder=MaxentClassifier.train and only got to 98.4% accuracy. So I can only conclude that a different feature detector is used. Hopefully the NLTK leaders will publish the training method so we can all know for sure.

Efficiency

On the nltk-users list, there was a question about which tagger is the most computationaly economic. I can't tell you the right answer, but I can definitely say that ClassifierBasedPOSTagger is the wrong answer. During accuracy evaluation, I noticed that the cpos tagger took a lot longer than raubt or braubt. So I ran timeit on the tag method of each tagger, and got the following results:

This was run with python 2.6.4 on an Athlon 64 Dual Core 4600+ with 3G RAM, but the important thing is the relative times. braubt is over 246 times faster than cpos! To put it another way, braubt can process over 66666 words/sec, where cpos can only do 270 words/sec and postag only 483 words/sec. So the lesson is: do not use a classifier based tagger if speed is an issue.

Here's the code for timing postag. You can do the same thing for any other pickled tagger by replacing nltk.tag._POS_TAGGER with a nltk.data accessible path with a .pickle suffix for the load method.

import nltk, timeit
text = nltk.word_tokenize('And now for something completely different')
setup = 'import nltk.data, nltk.tag; tagger = nltk.data.load(nltk.tag._POS_TAGGER)'
t = timeit.Timer('tagger.tag(%s)' % text, setup)
print 'timing postag 1000 times'
spent = t.timeit(number=1000)
print 'took %.5f secs/pass' % (spent / 1000)

File Size

There's also a significant difference in the file size of the pickled taggers (trained on treebank):

Fin

I think there's a lot of room for experimentation with classifier based taggers and their feature detectors. But if speed is an issue for you, don't even bother. In that case, stick with a simpler tagger that's nearly as accurate and orders of magnitude faster.

import logging

class InfoFilter(logging.Filter):
	def filter(self, rec):
		return rec.levelno == logging.INFO

Configuring Python Logging Filters

Filters can be added to a logger instance or a handler instance using the addFilter(filt) method. For a logger, the best time to do this is probably right after calling getLogger, like so:

log = logging.getLogger()
log.addFilter(InfoFilter())

What about adding a filter to a handler? If you're programmatically configuring handlers with addHandler(hdlr), then you can do the same thing by calling addFilter(filt) on the handler instance. But if you're using fileConfig to configure handlers and loggers, it's a little bit harder. Unfortunately, the logging configuration format does not support adding filters. And it's not always clear which logger the handler instances are attached to in the logger hierarchy. So the simplest way to add a filter to a handler in this case is to subclass the handler:

class InfoHandler(logging.StreamHandler):
	def __init__(self, *args, **kwargs):
		StreamHandler.__init__(self, *args, **kwargs)
		self.addFilter(InfoFilter())

Then in your file config, make sure to set the class value for your custom handler to a complete code path for import:

[handler_infohandler]
class=mypackage.mylogging.InfoHandler
level=INFO

Now your handler will only handle the log records that pass your custom filter. As long your handlers aren't changing much, the above method is much more reusable than having to call addFilter(filt) everytime a new logger is instantiated.

• Python Unicode – Fixing UTF-8 encoded as Latin-1 / ISO-8859-1
• Unicode HOWTO
• Universal Encoding Detector