Training Tree Transducers

Many probabilistic models for natural language are now written in terms of hierarchical tree structure. Tree-based modeling still lacks many of the standard tools taken for granted in (finitestate) string-based modeling. The theory of tree transducer automata provides a possible framework to draw on, as it has been worked out in an extensive literature. We motivate the use of tree transducers for natural language and address the training problem for probabilistic tree-totree and tree-to-string transducers.

• Type: Conference
• Year: 2004
• Where: NAACL
• Authors: Jonathan Graehl, Kevin Knight
• Keywords: Many Probabilistic Models, NAACL 2004, NAACL 2007, Natural Language, Tree Transducers