Abstract
We present a novel method for the computation of so-called prefix probabilities for linear context-free rewriting systems. Our technique streamlines previous procedures to compute prefix probabilities for probabilistic context-free grammars, probabilistic synchronous context-free grammars and probabilistic tree adjoining grammars. In addition, the methodology is general enough to be used for a wider range of problems involving, for example, several prefixes.
| Original language | English |
|---|---|
| Pages (from-to) | 331-350 |
| Number of pages | 33 |
| Journal | Journal of Logic and Computation |
| Volume | 24 |
| Issue number | 2 |
| Early online date | 23 Jan 2013 |
| DOIs | |
| Publication status | Published - Apr 2014 |
Keywords
- Linear context-free rewriting systems
- Language models
- Grammar transformations
- Probablistic grammars
- Prefix probability
