Count-based state merging for probabilistic regular tree grammars
Abstract
We present an approach to obtain language models from a tree corpus using probabilistic regular tree grammars (prtg). Starting with a prtg only generating trees from the corpus, the prtg is generalized step by step by merging nonterminals. We focus on bottom-up deterministic prtg to simplify the calculations.
Citation
Dietze , T & Nederhof , M J 2015 , Count-based state merging for probabilistic regular tree grammars . in Proceedings of the 12th International Conference on Finite State Methods and Natural Language Processing . Association for Computational Linguistics , Duesseldorf, Germany , The 12th International Conference on Finite-State Methods and Natural Language Processing , Düsseldorf , Germany , 22/06/15 . conference
Publication
Proceedings of the 12th International Conference on Finite State Methods and Natural Language Processing
Type
Conference item
Rights
Copyright ACL 2015. Licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 License.
Collections
Items in the St Andrews Research Repository are protected by copyright, with all rights reserved, unless otherwise indicated.