Table 5.

Performance metrics of machine learning models for RCC risk prediction (training set).

Genetic model	Model	Accuracy	Sensitivity	Specificity	Precision	Recall	F 1
Codominant	Logistic regression	0.5170	0.5519	0.4826	0.5120	0.5519	0.5312
	Elastic net	0.5480	0.5344	0.5628	0.5725	0.5344	0.5528
	C5.0 Decision Tree	0.8547	0.8799	0.8264	0.8502	0.8799	0.8648
	Random Forests	0.9114	0.9021	0.9207	0.9192	0.9021	0.9106
	Radial kernel SVM	0.3483	0.3753	0.3229	0.3435	0.3753	0.3587
Dominant	Logistic regression	0.6563	0.6988	0.6166	0.6303	0.6988	0.6628
	Elastic net	0.7066	0.7425	0.6719	0.6867	0.7425	0.7135
	C5.0 Decision Tree	0.8823	0.9149	0.8493	0.8600	0.9149	0.8866
	Random Forests	0.9172	0.9021	0.9324	0.9303	0.9021	0.9160
	Radial kernel SVM	0.5767	0.5714	0.5823	0.5933	0.5714	0.5822
Recessive	Logistic regression	0.6754	0.6810	0.6697	0.6749	0.6810	0.6779
	Elastic net	0.5871	0.5516	0.6224	0.5919	0.5516	0.5711
	C5.0 Decision Tree	0.8384	0.8758	0.8009	0.8151	0.8758	0.8444
	Random Forests	0.9196	0.9068	0.9324	0.9306	0.9068	0.9185
	Radial kernel SVM	0.5222	0.5034	0.5405	0.5150	0.5034	0.5091