Latent-variable modeling of string transductions with finite-state methods
Citations Over TimeTop 10% of 2008 papers
Abstract
String-to-string transduction is a central problem in computational linguistics and natural language processing. It occurs in tasks as diverse as name transliteration, spelling correction, pronunciation modeling and inflectional morphology. We present a conditional loglinear model for string-to-string transduction, which employs overlapping features over latent alignment sequences, and which learns latent classes and latent string pair regions from incomplete training data. We evaluate our approach on morphological tasks and demonstrate that latent variables can dramatically improve results, even when trained on small data sets. On the task of generating morphological forms, we outperform a baseline method reducing the error rate by up to 48%. On a lemmatization task, we reduce the error rates in Wicentowski (2002) by 38--92%.
Related Papers
- → Latent Variable Models(2004)619 cited
- → Modeling Configural Patterns in Latent Variable Profiles: Association With an Endogenous Variable(2014)8 cited
- → Latent Variable Models(2016)4 cited
- Nonlinear Effects in the Generalized Latent Variable Model(2006)
- → A Multilevel Latent Variable Model for Multidimensional Longitudinal Data(2009)