NLP | Brill Tagger

• BrillTagger class is a transformation-based tagger. It is not a subclass of SequentialBackoffTagger.
• Moreover, it uses a series of rules to correct the results of an initial tagger.
• These rules it follows are scored based. This score is equal to the no. of errors they correct minus the no. of new errors they produce.

Code #1 : Training a BrillTagger class

# Loading Libraries
from nltk.tag import brill, brill_trainer

def train_brill_tagger(initial_tagger, train_sents, **kwargs):
    templates = [
            brill.Template(brill.Pos([-1])),
            brill.Template(brill.Pos([1])),
            brill.Template(brill.Pos([-2])),
            brill.Template(brill.Pos([2])),
            brill.Template(brill.Pos([-2, -1])),
            brill.Template(brill.Pos([1, 2])),
            brill.Template(brill.Pos([-3, -2, -1])),
            brill.Template(brill.Pos([1, 2, 3])),
            brill.Template(brill.Pos([-1]), brill.Pos([1])),
            brill.Template(brill.Word([-1])),
            brill.Template(brill.Word([1])),
            brill.Template(brill.Word([-2])),
            brill.Template(brill.Word([2])),
            brill.Template(brill.Word([-2, -1])),
            brill.Template(brill.Word([1, 2])),
            brill.Template(brill.Word([-3, -2, -1])),
            brill.Template(brill.Word([1, 2, 3])),
            brill.Template(brill.Word([-1]), brill.Word([1])),
            ]

    # Using BrillTaggerTrainer to train
    trainer = brill_trainer.BrillTaggerTrainer(
            initial_tagger, templates, deterministic = True)

    return trainer.train(train_sents, **kwargs)

Code #2 : Let’s use the trained BrillTagger

from nltk.tag import brill, brill_trainer
from nltk.tag import DefaultTagger
from nltk.corpus import treebank
from tag_util import train_brill_tagger

# Initializing
default_tag = DefaultTagger('NN')

# initializing training and testing set
train_data = treebank.tagged_sents()[:3000]
test_data = treebank.tagged_sents()[3000:]

initial_tag = backoff_tagger(
        train_data, [UnigramTagger, BigramTagger,
                    TrigramTagger], backoff = default_tagger)

a = initial_tag.evaluate(test_data)
print ("Accuracy of Initial Tag : ", a)

Output :

Accuracy of Initial Tag : 0.8806820634578028

Code #3 :

brill_tag = train_brill_tagger(initial_tag, train_data)
b = brill_tag.evaluate(test_data)

print ("Accuracy of brill_tag : ", b)

Output :

Accuracy of brill_tag : 0.8827541549751781