Skip to main content
. 2009 Oct 5;106(41):17284–17289. doi: 10.1073/pnas.0905638106

Table 2.

Mean recall (and perplexity) when parsing different test sets with different grammars

Brian 2;0 grammar Annie 2;0 grammar Brian 3;0 grammar Annie 3;0 grammar
Brian 2;0 utterances 84% (105.4) 36% (636.3) 46% (1,076) 34% (1,486.2)
Annie 2;0 utterances 15% (381.9) 75% (184.1) 71% (317.6) 81% (425.9)
Brian 3;0 utterances 8% (455.7) 42% (361.5) 70% (364.6) 63% (363.7)
Annie 3;0 utterances 3% (489.5) 29% (526.4) 59% (575.8) 81% (276.5)