Ensemble Methods with Voting Protocols Exhibit Superior Performance for Predicting Cancer Clinical Endpoints and Providing More Complete Coverage of Disease-Related Genes
Table 5
Voting results of OV.
Platforms
Class
Vote from all methods
Vote after filtering
MCC avg
MCC max
OV-mRNA
bayes
0.4725
0.4725
0.1217
0.7906
functions
0.1250
0.4725
−0.0890
0.7906
lazy
0.3162
0.3162
0.1328
0.6447
meta
0.1890
0.1890
0.1923
0.6447
misc
0.6139
0.7500
0.3433
0.8864
rules
0.1250
0.3162
−0.1160
0.8864
trees
0.1890
0.1890
0.1734
0.6447
overall
0.3162
0.3162
0.0890
0.8864
OV-miRNA
bayes
1.0000
0.8771
0.2508
0.8864
functions
0.5000
0.6139
−0.0740
0.8771
lazy
0.6447
0.6447
0.2756
0.6325
meta
0.7559
0.7559
0.3786
0.8864
misc
0.7559
0.6139
0.1384
0.6447
rules
0.3430
0.7559
−0.0783
0.6447
trees
0.5000
0.7559
0.1258
1.0000
overall
0.7559
0.7559
0.1432
1.0000
OV-mRNA And OV-miRNA
bayes
0.4725
0.4725
0.1863
0.8864
functions
−1.0000
0.7500
−0.0815
0.8771
lazy
0.7500
0.8771
0.2042
0.6447
meta
0.3162
0.3162
0.2854
0.8864
misc
0.6139
0.7500
0.2408
0.8864
rules
0.3430
0.3162
−0.0971
0.8864
trees
0.3162
0.4725
0.1496
1.0000
overall
0.3162
0.4725
0.1161
1.0000
All of the measurements in the tables are MCCs, and the vote after filtering is the MCC based on the eliminated methods. The “avg” is the average of the MCCs.