BayesianPOSTagger

A Bayesian POS Tagger in Perl 5. Algorithms implemented:

1. Naive Bayes
2. Complete Naive Bayes (CNB)
3. Gibbs Sampling: random extraction, max frequent on to_sample samples and max on position

Idea and algorithm described in the following papers:

1. For CNB: Part of Speech Tagging with Naïve Bayes Methods
2. For Gibbs Sampling(Idea for the approach): Bayesian Analysis for Natural Language Processing

Dataset

The dataset used is the Brown's corpus. You can find it here. The meaning of the tags can be found here.