Skip to Main Content
Table 8

Chinese test set results (PCTB articles 271–300) applying sentence-level high-precision and unary constraints to two parsers with parameter settings tuned on development data.

Parser
F1
Precision
Recall
Seconds
Speed-up
BUBS (2010) 79.5 79.5 79.1 169  
 + Unary(50) 80.7 82.1 79.4 153 1.1x 
 + HP(0.8) 81.1 81.5 80.7 75 2.3x 
 + HP(0.8) + Unary(50) 81.8 83.1 80.5 44 3.8x 
 
Berkeley (2007) 83.9 84.5 83.3 141  
 + Unary(50) 84.5 85.9 83.0 125 1.1x 
 + HP(0.7) 84.5 85.1 83.8 64 2.2x 
 + HP(0.7) + Unary(50) 84.7 86.1 83.4 57 2.5x 
Parser
F1
Precision
Recall
Seconds
Speed-up
BUBS (2010) 79.5 79.5 79.1 169  
 + Unary(50) 80.7 82.1 79.4 153 1.1x 
 + HP(0.8) 81.1 81.5 80.7 75 2.3x 
 + HP(0.8) + Unary(50) 81.8 83.1 80.5 44 3.8x 
 
Berkeley (2007) 83.9 84.5 83.3 141  
 + Unary(50) 84.5 85.9 83.0 125 1.1x 
 + HP(0.7) 84.5 85.1 83.8 64 2.2x 
 + HP(0.7) + Unary(50) 84.7 86.1 83.4 57 2.5x 
Close Modal

or Create an Account

Close Modal
Close Modal