I am self-reading the NLP book by Dan Jurafsky. In the book, the following equation is given for incorporating the sentence end marker in the Viterbi algorithm for POS tagging.

enter image description here

I am confused why the sentence end marker is treated specially to use bigram, instead of trigram as usual. Why is the sentence start marker not treated this way?

  • Where in the book are you referencing? My $0.02. Ngrams are used for a variety of needs depending upon the depth of search needed, available computational resources, data variables (5V's) and more. Could this algorithm be more specificly tuned to, or designed for bigrams vs trigrams? – CoffeeBaconAddict Dec 30 '20 at 3:34

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.