roopalgarg
roopalgarg

Reputation: 449

Discriminatively trained supervised part of speech tagging

I have to implement a Discriminatively trained supervised part of speech tagger, and I have been looking at a couple of techniques including Maximum likelihood, perceptron and the large margin (SVM). Finally after reading through some experimental results quoted in a couple of research papers i have come down to using SVMs for it. I have been studying it for some time and a couple of things in theory seem a little confusing. Can someone please point me to some relevant reading material to a practical implementation or just more clarification on how to implement it using Viterbi Algorithm.

P.S. : I am not asking for the solution, but just need some guidance.

Upvotes: 0

Views: 116

Answers (1)

oroszgy
oroszgy

Reputation: 123

I would suggest, to read some classic papers.

And a step-by-step construction paper using Maxent in Python NLTK:

You could also study some open source software, such as Apache OpenNLP, Python NLTK or my own implementation PurePos

For using SVM, you could look around here and here.

Upvotes: 1

Related Questions