Classification of Cancer Primary Sites Using Machine Learning and Somatic Mutations
Table 3
Micro- and macroaveraged accuracies of seven combinations of gene symbols with three other features.
Feature combination
Number of features
miAccuracy
maAccuracy (mean)
maAccuracy (SD)
Gene (baseline)
21,286
0.57
0.57
0.019
Gene + gMutation
101,151
0.58
0.58
0.019
Gene + Pathway
21,571
0.58
0.58
0.010
Gene + Chromosome
21,311
0.60
0.60
0.022
Gene + gMutation + Pathway
101,436
0.60
0.60
0.013
Gene + gMutation + Chromosome
101,176
0.62
0.62
0.021
Gene + gMutation + Chromosome + Pathway
101,461
0.60
0.60
0.015
Note: miAccuracy represents the microaverage accuracy; maAccuracy represents the macroaverage accuracy, which is reported in mean and standard deviation (SD) over 10 accuracies from 10-fold cross validation.
