Table 4

Gold Standard: 0/1-loss over all microlabel predictions with GTG kernel combined with linear or degree 51 polynomial kernel.

Sequence Kernel

Nearest neighbor (std)

MMR linear (std)

MMR poly-51 (std)

HM3 linear (std)

HM3 poly-51


GTG

24.1 (1.9)

36.3 (2.8)

21.4 (1.8)

23.3 (2.0)

21.6 (1.7)


Astikainen et al. BMC Proceedings 2008 2(Suppl 4):S2  

Open Data