School of somethingFACULTY OF OTHER

School of ComputingFACULTY OF ENGINEERING

Machine Learning PoS-Taggers

COMP3310 Natural Language Processing

Eric Atwell, Language Research Group

(with thanks to Katja Markert, Marti Hearst, and other contributors)

Reminder

Puns play on our assumptions of the next word…

… eg they present us with an unexpected homonym (bends)

ConditionalFreqDist() counts word-pairs: word bigrams

Used for story generation, Speech recognition, …

Parts of Speech: groups words into grammatical categories

… and separates different functions of a word

In English, many words are ambiguous: 2 or more PoS-tags

Very simple tagger: tag with the likeliest tag for the word

Better Pos-Taggers: to come…

Taking context into account

Theory behind some example Machine Learning PoS-taggers

Example implementations in NLTK

Machine Learning from a PoS-tagged training corpus

Statistical (N-Gram/Markov) taggers:

learn table of 1/2/3/N-tag sequence frequencies

Brill (transformation-based) tagger:

learn likeliest tag for each word ignoring context,

then learn rules to change tag to fit context

NB you don’t have to use NLTK – just useful to illustrate

Training and Testing ofMachine Learning Algorithms

Algorithms that “learn” from data see a set of examples and try to generalize from them.

Training set:

• Examples trained on

Test set:

• Also called held-out data and unseen data

• Use this for evaluating your algorithm

• Must be separate from the training set; otherwise, you cheated!

“Gold standard” evaluation corpus

• An evaluation set that a community has agreed on and uses as a common benchmark.

• Not “seen” until development is finished – ONLY for evaluation

Cross-Validation of Learning Algorithms

Cross-validation set

• Part of the training set.

Used for tuning parameters of the algorithm without “polluting” (tuning to) the test data.

• You can train on x%, and then cross-validate on the remaining 1-x%

• E.g., train on 90% of the training data, cross-validate (test) on the remaining 10%

• Repeat several times with different splits

• This allows you to choose the best settings to then use on the real test set.

• You should only evaluate on the test set at the very end, after you’ve gotten your algorithm as good as possible on the cross-validation set.

Strong Baselines

When designing NLP algorithms, you need to evaluate them by comparing to others.

Baseline Algorithm:

• An algorithm that is relatively simple but can be expected to do “ok”

• Should get the best score possible by doing the obvious thing.

A Tagging Baseline

Find the most likely tag for the most frequent words

• Frequent words are ambiguous

• You’re likely to see frequent words in any collection

• Will always see “to” but might not see “armadillo”

How to do this?

• First find the most likely words and their tags in the training data

• Train a tagger that looks up these results in a table

Find the most frequent words and the most likely tag of each

Use our own tagger class

N-Grams

The N stands for how many terms are used

• Unigram: 1 term (0th order)

• Bigram: 2 terms (1st order)

• Trigrams: 3 terms (2nd order)

• Usually don’t go beyond this

You can use different kinds of terms, e.g.:

• Character based n-grams

• Word-based n-grams

• POS-based n-grams

Ordering

• Often adjacent, but not required

We use n-grams to help determine the context in which some linguistic phenomenon happens.

E.g., look at the words before and after period to see if it is the end of sentence or not.

Tagging with lexical frequencies

Secretariat/NNP is/VBZ expected/VBN to/TO race/VB tomorrow/NN

People/NNS continue/VBP to/TO inquire/VB the/DT reason/NN for/IN the/DT race/NN for/IN outer/JJ space/NN

Problem: assign a tag to race given its lexical frequency

Solution: we choose the tag that has the greater probability

• P(race|VB)

• P(race|NN)

Unigram Tagger

Train on a set of sentences

Keep track of how many times each word is seen with each tag.

After training, associate with each word its most likely tag.

• Problem: many words never seen in the training data.

• Solution: have a default tag to “backoff” to.

Unigram tagger with Backoff

What’s wrong with unigram?

Most frequent tag isn’t always right!

Need to take the context into account

• Which sense of “to” is being used?

• Which sense of “like” is being used?

N-gram tagger

Uses the preceding N-1 predicted tags

Also uses the unigram estimate for the current word

Bigram Tagging

• For tagging, in addition to considering the token’s type, the context also considers the tags of the n preceding tokens

• What is the most likely tag for word n, given word n-1 and tag n-1?

• The tagger picks the tag which is most likely for that context.

Combining Taggers using Backoff

Use more accurate algorithms when we can, backoff to wider coverage when needed.

• Try tagging the token with the 1st order tagger.

• If the 1st order tagger is unable to find a tag for the token, try finding a tag with the 0th order tagger.

• If the 0th order tagger is also unable to find a tag, use the default tagger to find a tag.

Important point:

• Bigram and trigram taggers use the previous tag context to assign new tags. If they see a tag of “None” in the previous context, they will output “None” too.

Demonstrating the n-gram taggers

Trained on brown.tagged(‘a’), tested on brown.tagged(‘b’)

Backs off to a default of ‘nn’

Demonstrating the n-gram taggers

Combining Taggers

The bigram backoff tagger did worse than the unigram! Why?

Why does it get better again with trigrams?

How can we improve these scores?

Rule-Based Tagger

The Linguistic Complaint

• Where is the linguistic knowledge of a tagger?

• Just a massive table of numbers

• Aren’t there any linguistic insights that could emerge from the data?

• Could thus use handcrafted sets of rules to tag input sentences, for example, if input follows a determiner tag it as a noun.

• Constraint Grammar (CG) tagger: PhD student spends 3+ years coding a large set of these rules (for English, Finnish, …)

• Machine Learning researchers would prefer to use ML to extract a large set of such rules from a PoS-tagged training corpus

The Brill tagger

An example of Transformation-Based Learning

• Basic idea: do a quick job first (using frequency), then revise it using contextual rules.

Very popular (freely available, works fairly well)

A supervised method: requires a tagged corpus

Brill Tagging: In more detail

Start with simple (less accurate) rules…learn better ones from tagged corpus

• Tag each word initially with most likely POS

• Examine set of transformations to see which improves tagging decisions compared to tagged corpus

• Re-tag corpus using best transformation

• Repeat until, e.g., performance doesn’t improve

• Result: tagging procedure (ordered list of transformations) which can be applied to new, untagged text

An example

Examples:

• They are expected to race tomorrow.

• The race for outer space.

Tagging algorithm:

1. Tag all uses of “race” as NN (most likely tag in the Brown corpus)

• They are expected to race/NN tomorrow

• the race/NN for outer space

2. Use a transformation rule to replace the tag NN with VB for all uses of “race” preceded by the tag TO:

• They are expected to race/VB tomorrow

• the race/NN for outer space

Example Rule Transformations

Sample Final Rules

Summary: N-gram/Markov and Transformation/Brill PoS-Taggers

Theory behind some example Machine Learning PoS-taggers

Example implementations in NLTK

Machine Learning from a PoS-tagged training corpus

Statistical (N-Gram/Markov) taggers:

learn table of 1/2/3/N-tag sequence frequencies

If not enough data for N, back off to N-1 patterns

Brill (transformation-based) tagger:

learn likeliest tag for each word ignoring context,

then learn rules to change tag to fit context

NB you don’t have to use NLTK – just useful to illustrate