Scientific journal NRU ITMO
Series "Processes and Food Production Equipment"
Registration certificate ЭЛ № ФС 77 – 55245 от 04.09.2013
зарегистрировано в Федеральной службе по надзору в сфере связи, информационных технологий и массовых коммуникаций

June 2014 (published: 01.06.2014)

Number 2(20)

Home > Issue > Modeling a modern POS tagger using HMM and Viterbi Algorithm

UDC 541.628

Modeling a modern POS tagger using HMM and Viterbi Algorithm

Shatornaya A., Vorobiev A.

The paper presents a modern way of an improved POS tagger which was modeled by using Hidden Markov Model and Viterbi Algorithm. The algorithm provides ideas for handling unknown words. This issue is caused by such words that have the same spelling but relate to complete different parts of speech and have various meanings. The accuracy of such POS tagger exceeds 96%. Such algorithm’s accuracy was analyzed to other existing algorithms and suggestions were provided.
Read the full article 

Keywords: HMM, Viterbi Algorithm, POS tagger, part of speech tagging

Editorial address:
191002, St. Petersburg, street. Lomonosova 9, office. 2132

Information 2006-2020, All rights reserved
Development by Department of Internet Solutions NRU ITMO © 2013