A probabilistic earley parser as a psycholinguistic model
NAACL '01 Proceedings of the second meeting of the North American Chapter of the Association for Computational Linguistics on Language technologies
Trust Region Newton Method for Logistic Regression
The Journal of Machine Learning Research
Surprising parser actions and reading difficulty
HLT-Short '08 Proceedings of the 46th Annual Meeting of the Association for Computational Linguistics on Human Language Technologies: Short Papers
Incrementality in deterministic dependency parsing
IncrementParsing '04 Proceedings of the Workshop on Incremental Parsing: Bringing Engineering and Cognition Together
Hi-index | 0.00 |
This paper examines how grammatical and memory constraints explain gradience in superiority violation acceptability. A computational model encoding both categories of constraints is compared to experimental evidence. By formalizing memory capacity as beam-search in the parser, the model predicts gradience evident in human data. To predict attachment behavior, the parser must be sensitive to the types of nominal intervenors that occur between a wh-filler and its head. The results suggest memory is more informative for modeling violation gradience patterns than grammatical constraints.