next up previous contents
Next: The Grammar Development Environment Up: System Description Previous: Tree Database

Tree Grafting

Once a particular set of lexicalized trees for the sentence have been selected, XTAG uses an Earley-style predictive left-to-right parsing algorithm for LTAGs ([#!schabesjoshi88!#,#!schabes90!#]) to find all derivations for the sentence. The derivation trees and the associated derived trees can be viewed using the X-interface (see Table 3.1). The X-interface can also be used to save particular derivations to disk. The output of the parser for the sentence I had a map yesterday is illustrated in Figure 3.2. The parse tree3.3 represents the surface constituent structure, while the derivation tree represents the derivation history of the parse. The nodes of the derivation tree are the tree names anchored by the lexical items3.4. The composition operation is indicated by the nature of the arcs: a dashed line is used for substitution and a bold line for adjunction. The number beside each tree name is the address of the node at which the operation took place. The derivation tree can also be interpreted as a dependency graph with unlabeled arcs between words of the sentence.
  
Figure 3.2: Output Structures from the Parser
\includegraphics[height=3.0in]{/mnt/linc/xtag/work/doc/tech-rept/ps/overview-files/derived.ps} \includegraphics[height=2.0in,width=2.7in]{/mnt/linc/xtag/work/doc/tech-rept/ps/overview-files/derivation.ps}
Parse Tree Derivation Tree


 
Table 3.3: XTAG tagset
Part of Speech Description
A Adjective
Ad Adverb
Comp Complementizer
D Determiner
G Genitive Noun
I Interjection
N Noun
P Preposition
PL Particle
Punct Punctuation
V Verb
 


next up previous contents
Next: The Grammar Development Environment Up: System Description Previous: Tree Database
XTAG Project
1998-09-14