Serial Combination of Rules and Statistics: A Case Study in Czech Tagging (0)
by
Jan Hajic
,
Pavel Krbec
,
Pavel Kveton
,
Karel Oliva
,
Vladimír Petkevič
| Citations: | 24 - 0 self |
BibTeX
@MISC{Hajic_serialcombination,
author = {Jan Hajic and Pavel Krbec and Pavel Kveton and Karel Oliva and Vladimír Petkevič},
title = {Serial Combination of Rules and Statistics: A Case Study in Czech Tagging},
year = {}
}
Years of Citing Articles
Bookmark
 |
 |
 |
 |
 |
 |
OpenURL
Abstract
A hybrid system is described which combines the strength of manual rulewriting and statistical learning, obtaining results superior to both methods if applied separately. The combination of a rule-based system and a statistical one is not parallel but serial: the rule-based system performing partial disambiguation with recall close to 100% is applied first, and a trigram HMM tagger runs on its results. An experiment in Czech tagging has been performed with encouraging results.
