P03-1055 unlexicalized PCFG augmented with the antecedent recovery algorithm . We use an unlexicalized
W03-1005 variables , which further complicates antecedent recovery , even in the case of perfect
P03-1055 encoding necessary information for antecedent recovery , do not incur any substantial
P03-1055 invaluable for EE detection and antecedent recovery . Paradoxically , taking the
P03-1055 interest lies in trace detection and antecedent recovery , we adopt the evaluation measures
W03-1005 Table 4 gives a closer look at the antecedent recovery score for some common EE types
P03-1055 overall precision and recall of antecedent recovery is 91.4 % . 3.4 Discussion The
W03-1005 the Sloan Foundation . <title> Antecedent Recovery : Experiments with a Trace Tagger
P03-1055 phrase . In order to facilitate antecedent recovery and to disambiguate the EEs ,
W03-1005 match neither the accuracy for antecedent recovery nor the time efficiency of the
P03-1055 efficiently and successfully tackle antecedent recovery . 5 Combining the models In Section
P04-1040 Dienes and Dubey , 2003 ) for antecedent recovery using Collins ' parser . Interestingly
W03-1005 of parsing , EE detection and antecedent recovery , but also with the running time
P03-1055 trace tagger , and comparing the antecedent recovery accuracy to Johnson ( 2002 )
W03-1005 step is beneficial to both to antecedent recovery accuracy and to parsing efficiency
P03-1055 the difference in speed and in antecedent recovery accuracy between the parser that
P03-1055 and after it ) . When evaluating antecedent recovery , the EEs are regarded as four-tuples
P02-1018 that can evaluate empty node and antecedent recovery , as described in section 3 .
W03-1005 case , PERFECT , the F-score for antecedent recovery is quite high in both the unlexicalized
W03-1005 , we compare EE detection and antecedent recovery F-scores of the TAGGER and the
hide detail