Sciweavers

Free Online Productivity Tools i2Speak i2Symbol i2OCR iTex2Img iWeb2Print iWeb2Shot i2Type iPdf2Split iPdf2Merge i2Bopomofo i2Arabic i2Style i2Image i2PDF iLatex2Rtf Sci2ools

15

ACL
1997

favoriteEmaildiscussreport

98views Computational Linguistics» more ACL 1997»

Finite State Transducers Approximating Hidden Markov Models

13 years 5 months ago

Finite State Transducers Approximating Hidden Markov Models

Download acl.ldc.upenn.edu

This paper describes the conversion of a Hidden Markov Model into a sequential transducer that closely approximates the behavior of the stochastic model. This transformation is especially advantageous for part-of-speech tagging because the resulting transducer can be composed with other transducers that encode correction rules for the most frequent tagging errors. The speed of tagging is also improved. The described methods have been implemented and successfully tested on six languages.

André Kempe

Real-time Traffic

ACL 1997 | ACL 2007 | Frequent Tagging Errors | Hidden Markov Model | Part-of-speech Tagging |

claim paper

Related Content

» PACLearning of Markov Models with Hidden State

» The Parametrized Probabilistic FiniteState Transducer Probe Game Player Fingerprint Model

» Concavity of Mutual Information Rate for InputRestricted FiniteState Memoryless Channels a...

» Inducing Hidden Markov Models to Model LongTerm Dependencies

» Approximating a nonhomogeneous HMM with Dynamic Spatial Dirichlet Process

» Large margin training for hidden Markov models with partially observed states

» Beam sampling for the infinite hidden Markov model

» The Infinite Hidden Markov Model

» Aggregationbased model reduction of a Hidden Markov Model

Post Info
More Details (n/a)

Added	01 Nov 2010
Updated	01 Nov 2010
Type	Conference
Year	1997
Where	ACL
Authors	André Kempe

Comments (0)