Graphical Models over Multiple Strings

We study graphical modeling in the case of stringvalued random variables. Whereas a weighted finite-state transducer can model the probabilistic relationship between two strings, we are interested in building up joint models of three or more strings. This is needed for inflectional paradigms in morphology, cognate modeling or language reconstruction, and multiple-string alignment. We propose a Markov Random Field in which each factor (potential function) is a weighted finite-state machine, typically a transducer that evaluates the relationship between just two of the strings. The full joint distribution is then a product of these factors. Though decoding is actually undecidable in general, we can still do efficient joint inference using approximate belief propagation; the necessary computations and messages are all finitestate. We demonstrate the methods by jointly predicting morphological forms. 1 Overview This paper considers what happens if a graphical model's variables can ra...

• Type: Journal
• Year: 2009
• Where: EMNLP
• Authors: Markus Dreyer, Jason Eisner
• Keywords: EMNLP 2009, Graphical Models, Natural Language Processing, Random Variables, Weighted Finite-state Transducer