Skip to main content
. 2010 Oct 11;1:24. doi: 10.4103/2153-3539.71065

Table 3.

Overall algorithm performance with both frame methods. Averaged algorithm performance on both questions versus the reference standard

Dynamic-Window ConText
Accuracy 0.962 0.962
Precision 0.893 0.877
Recall 0.944 0.961
F-measure 0.918 0.916

Avg kappa (ref std) 0.928